JP2009119958A - Vehicle state estimation unit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vehicle state estimation unit for estimating a state amount of a vehicle in an all wheel drive vehicle, which easily estimates the vehicle state amount with high accuracy, by estimating the vehicle state amount obtained by detecting the vehicle state using as small number of sensors as possible. <P>SOLUTION: In S8, an estimated vehicle speed V<SB>est</SB>is calculated from an average wheel speed V<SB>wave</SB>of the vehicle and an equivalent slip ratio λ<SB>avg</SB>. The estimated vehicle speed V<SB>est</SB>during acceleration is calculated by V<SB>est</SB>=(1-slip ratio λ<SB>avg</SB>)×V<SB>wave</SB>. The estimated vehicle speed V<SB>est</SB>during deceleration is calculated by V<SB>est</SB>=V<SB>wave</SB>/(1+slip ratio λ<SB>avg</SB>). By calculating the estimated vehicle speeds V<SB>est</SB>in the above manner, the estimated vehicle speeds V<SB>est</SB>during acceleration and deceleration can be calculated properly. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a vehicle state estimation device that estimates a vehicle state quantity, and more particularly to a vehicle state estimation apparatus that estimates a vehicle state quantity such as a vehicle body speed in an all-wheel (four-wheel) drive vehicle.

  Conventionally, as a vehicle state estimation device for estimating a vehicle state quantity such as a vehicle body speed in an all-wheel drive vehicle, for example, estimation devices described in Patent Literature 1 and Patent Literature 2 below are known.

  Patent Document 1 discloses an estimation device that verifies a vehicle state quantity calculated by a predetermined equation of motion with a measurable state quantity and corrects an initial value for calculating the vehicle state quantity according to the verification result. FIG. 11 of Patent Document 1 discloses a block diagram of a vehicle body speed estimation device for estimating the vehicle body speed.

  Further, in Patent Document 2, the estimated vehicle body longitudinal direction estimated wheel speed is obtained from the estimated longitudinal acceleration, the vehicle body lateral direction estimated wheel speed is obtained from the estimated lateral acceleration, and the estimated vehicle body speed is estimated from the estimated longitudinal acceleration and the estimated lateral acceleration. An apparatus is disclosed.

JP 2003-306092 A JP 2003-118559 A

  As described above, in an all-wheel drive vehicle, since all the wheels are driven, it is difficult to detect the vehicle state quantity such as the vehicle body speed. Therefore, the vehicle state quantity is calculated based on the estimated value. There was a need.

  For this reason, in the above-mentioned patent documents, for example, as shown in FIG. 11 of Patent Document 1, in order to estimate the vehicle body speed in more detail, many expensive sensors such as a yaw rate sensor and a lateral acceleration sensor are provided. The calculation accuracy of the state quantity was improved.

However, when a large number of expensive sensors are provided, there arises a problem that the cost of the apparatus itself increases, and peripheral parts such as a harness for installing the sensor are also required, resulting in a problem that the apparatus itself becomes complicated.
In addition, if a large number of sensors are provided, an input signal to be input to the control device also increases, which causes a problem that the calculation processing for calculating the state quantity of the vehicle becomes complicated.

  Accordingly, the present invention provides a vehicle state estimation device that estimates a vehicle state quantity in an all-wheel-drive vehicle, detects the vehicle state by using as few sensors as possible, and estimates the vehicle state quantity. It is an object of the present invention to provide a vehicle state estimation device that can easily and accurately estimate the vehicle state.

  A vehicle state estimation device according to the present invention is a vehicle state estimation device for estimating a vehicle state quantity in an all-wheel drive vehicle, and detects longitudinal acceleration detection means for detecting longitudinal acceleration of a vehicle body and wheel speed of each wheel. The slip condition of each wheel is determined by the wheel speed detection means, the equivalent slip ratio calculation means for calculating the equivalent slip ratio of the slip state of all the wheels from the detected longitudinal acceleration, and the reference vehicle body speed and the wheel speed of each wheel. A wheel speed average calculating means for calculating an average wheel speed from a non-slip wheel whose slip state is a predetermined value or less, and an estimated vehicle speed calculating means for calculating an estimated vehicle speed from the equivalent slip ratio and the wheel speed average. Is.

According to the above configuration, by detecting the longitudinal acceleration and the wheel speed from the longitudinal acceleration detecting means and the wheel speed detecting means, the equivalent slip ratio of all the wheels is calculated, and the wheel speed average is calculated by the non-slip wheel. The estimated vehicle speed is calculated from the equivalent slip ratio and the average wheel speed.
For this reason, it is possible to calculate the estimated vehicle speed using only commonly used wheel speed detection means and longitudinal acceleration detection means without using expensive sensors such as a yaw rate sensor and a lateral acceleration sensor.
Further, since the wheel speed average is calculated for the non-slip wheels except for the slipping wheel, the wheel speed average can be calculated without taking in a wheel speed that is significantly different from the vehicle body speed. For this reason, the calculation accuracy of the estimated vehicle speed can be increased.
Here, the “equivalent slip ratio” is a term that means an average value of slip ratios of all wheels of all wheels, and is not obtained from an actual wheel slip ratio but is obtained from a longitudinal acceleration of a vehicle body (calculation). Value).

In one embodiment of the present invention, the reference vehicle body speed is calculated by adding a differential value of the previous estimated vehicle body speed and the previous longitudinal acceleration.
According to the above configuration, the reference vehicle speed for determining the slip state of the wheel is calculated by adding the previously calculated estimated vehicle speed and the differential value of the longitudinal acceleration detected last time. The reference vehicle speed can be obtained without providing a sensor or the like for detecting the reference vehicle speed.
Therefore, it is possible to obtain the reference vehicle speed by reducing the cost of the device itself without providing a sensor with poor feasibility.

In one embodiment of the present invention, the reference vehicle body speed is set to be calculated as the estimated vehicle body speed when all the wheels slip over a predetermined value.
According to the above configuration, when all the wheels slip to a predetermined value or more and there are no non-slip wheels, the average vehicle speed cannot be calculated by calculating the reference vehicle speed as the estimated vehicle speed as it is. Even in this case, the estimated vehicle speed can be calculated.
Therefore, the estimated vehicle body speed can be calculated even when the wheel speeds detected by the all-wheel slip are all greatly different from the vehicle body speed.

In one embodiment of the present invention, the calculated value of the equivalent slip ratio calculating means is changed according to the road surface μ.
According to the above configuration, the equivalent slip ratio of the wheel to be calculated changes according to the road surface μ, so that an equivalent slip ratio closer to the actually measured value can be calculated.
Therefore, it is possible to calculate the estimated vehicle speed reflecting the difference in the road surface μ.

In one embodiment of the present invention, the tire driving torque detecting means for detecting the driving torque for driving each wheel, and the wheel angle for calculating the angular acceleration of each wheel by differentiating the wheel speed detected by the wheel speed detecting means. From the acceleration calculation means, the estimated vehicle speed calculated by the estimated vehicle speed calculation means, the tire drive torque detected by the tire drive torque detection means, and the wheel angular acceleration calculated by the wheel angular acceleration calculation means, Tire braking / driving force estimating means for estimating braking / driving force is provided.
According to the above configuration, the tire braking / driving is calculated from the estimated vehicle speed calculated by the estimated vehicle speed calculating means, the tire driving torque detected by the tire driving torque detecting means, and the wheel angular acceleration calculated by the wheel angular acceleration calculating means. The tire braking / driving force generated by the tire is estimated by the force estimating means.
For this reason, it is possible to estimate the tire braking / driving force generated by each tire, which has been difficult to detect in the past, as accurately as possible. By using the estimated tire braking / driving force, the vehicle attitude control device Such control can be performed after predicting the behavior of the vehicle.
Therefore, it is possible to configure a vehicle control device that has previously estimated the tire braking / driving force that has been difficult to detect and has improved safety performance and the like.

In one embodiment of the present invention, the tire braking / driving force estimating means takes in at least an estimated vehicle body speed, a tire driving torque, and a wheel angular acceleration, and calculates an estimated value by comparing with a predetermined tire model. It is configured including a Kalman filter.
According to the above configuration, the tire braking / driving force estimating means includes the Kalman filter, so that the tire braking / driving force value is calculated by the estimated value of the Kalman filter.
For this reason, since the estimated value is not calculated from the actually detected detection value, the generation of noise of the tire braking / driving force, which is the calculated value, can be suppressed as much as possible, and the occurrence of control delay can also be prevented.
Therefore, the value of the tire braking / driving force can be calculated more accurately and quickly, and the performance of the vehicle control device can be improved by using the value of the tire braking / driving force.

According to this invention, it is possible to calculate the estimated vehicle body speed using only commonly used wheel speed detection means and longitudinal acceleration detection means without using an expensive sensor such as a yaw rate sensor or a lateral acceleration sensor.
Further, since the wheel speed average is calculated for the non-slip wheels except for the slipping wheel, the wheel speed average can be calculated without taking in a wheel speed that is significantly different from the vehicle body speed. For this reason, the calculation accuracy of the estimated vehicle speed can be increased.
Therefore, in the vehicle state estimation device that estimates the vehicle state quantity in the all-wheel drive vehicle, the vehicle state quantity is estimated by detecting the vehicle state by using as few sensors as possible, and thus the vehicle state quantity can be easily calculated. In addition, it can be estimated with high accuracy.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a schematic diagram of an automobile employing a vehicle state estimation device according to an embodiment of the present invention, FIG. 2 is a system block diagram of the vehicle state estimation device, and FIG. 3 is a vehicle speed estimation control method of the vehicle state estimation device. The first half of the flowchart to be described, FIG. 4 is the latter half of the flowchart for explaining the control method for vehicle speed estimation, FIG. 5 is a map for obtaining the equivalent slip ratio, and FIG. 6 is the flowchart for explaining the control method for tire braking / driving force estimation. is there.

As shown in FIG. 1, the automobile 1 is provided with a plurality of sensors and an actuator for a vehicle attitude control device.
First, as a sensor, a wheel speed sensor 3 for detecting the rotational speed of the wheel (hereinafter referred to as wheel speed) for each wheel 2... And a longitudinal G sensor 4 for detecting longitudinal acceleration (hereinafter referred to as longitudinal G) generated in the vehicle body. And a drive torque sensor 5 for detecting a drive torque transmitted to each wheel 2. When each wheel is driven by an electric motor as in an electric vehicle, the driving torque of each wheel can be calculated from the output torque of the electric motor.

  On the other hand, the actuator of the vehicle attitude control device includes a plurality of control units 7 that control the rotational torque of the electric motors 6 provided for each wheel.

  These sensors (3, 4, 5) and the control unit 7 are respectively connected to a vehicle body speed estimator 8 and a tire braking / driving force estimator 9 provided in the center as shown in the block diagram of FIG. The control unit is not shown).

  First, the front / rear G sensor 4 is connected to the acceleration filter unit 10 of the vehicle body speed estimator 8 and transmits a front / rear G signal to the acceleration filter unit 10.

  Further, the wheel speed sensor 3 provided for each wheel is connected to the wheel speed filter unit 11 of the vehicle body speed estimator 8 and transmits a wheel speed signal to the wheel speed filter unit 11. Further, the wheel speed signal of the wheel speed sensor 3 is branched in the middle and transmitted to the differential processing unit 14 and the tire braking / driving force filter unit 16 of the tire braking / driving force estimator 9.

  A driving torque sensor 5 provided for each wheel (axle) is connected to the tire braking / driving force calculation unit 15 and calculates a tire braking / driving force by using a driving torque signal that is a signal of the tire driving torque transmitted to each wheel. To the unit 15. As described above, when each wheel is driven by an electric motor, the driving torque of each wheel may be calculated by the control unit from the output torque of the motor.

On the other hand, output signals are transmitted from the vehicle attitude control device 17 to the actuator 7 of the vehicle attitude control device 17, although not shown.
The vehicle attitude control device 17 is connected to the vehicle body speed estimation unit 13 of the vehicle body speed estimator 8 and the tire braking / driving force filter unit 16 of the tire braking / driving force estimator 9. The estimated vehicle speed signal is received, and the estimated braking / driving force signal of each wheel is received from the tire braking / driving force filter unit 16.

  The block structure of the vehicle body speed estimator 8 and the tire braking / driving force estimator 9 will be described in more detail. First, the vehicle body speed estimator 8 includes an acceleration filter unit 10, a wheel speed filter unit 11, a memory unit 12, The vehicle body speed estimation unit 13 is provided. On the other hand, the tire braking / driving force estimator 9 includes a differential processing unit 14, a tire braking / driving force calculation unit 15, and a tire braking / driving force filter unit 16.

  The acceleration filter unit 10 of the vehicle body speed estimator 8 is composed of a low-pass filter so as to remove the high-frequency region of the front-rear G signal. By filtering the detected front-rear G signal, the front-rear G signal easy to process is obtained. Is generated. The front / rear G signal is transmitted to the vehicle body speed estimation unit 13.

  The wheel speed filter unit 11 is configured to capture the wheel speed of each wheel from each wheel speed sensor 3, and is configured to capture the previous estimated vehicle body speed and the previous front and rear G signal from the memory unit 12. Thus, the slip state (idling state or locked state) of each wheel is determined, and the wheel speed average is calculated from the non-slip wheel that is not slipping. The wheel speed average is transmitted to the vehicle body speed estimation unit 13.

  The memory unit 12 is configured to temporarily store the estimated vehicle body speed calculated during the previous control flow and the front and rear G detected during the previous control flow. In this control flow, each signal is transmitted to the wheel speed filter unit 11.

  As will be described later, the vehicle body speed estimation unit 13 is configured to calculate an estimated vehicle body speed using a predetermined slip ratio map determined by the tire braking / driving force and the slip ratio.

  As described above, the front-rear G signal filtered from the acceleration filter unit 10 is transmitted to the vehicle body speed estimation unit 13, and the wheel speed average is transmitted from the wheel speed filter unit 11. Further, a road surface μ signal for changing the slip ratio map is transmitted from the tire braking / driving force filter unit 16.

  The estimated vehicle body speed calculated by the vehicle body speed estimating unit 13 is output to the vehicle attitude control device 17 and the tire braking / driving force filter unit 16.

  The differential processing unit 14 of the tire braking / driving force estimator 9 takes in each wheel speed from each wheel speed sensor 3, calculates the angular acceleration of each wheel by differentiating the wheel speed, and calculates this angular acceleration to the tire control. It is configured to output to the driving force calculation unit 15.

  The tire braking / driving force calculation unit 15 is configured to calculate the tire braking / driving force of each wheel according to a tire braking / driving force calculation formula described later. The angular acceleration of each wheel is taken in from the differential processing unit 14 and the driving torque signal of each wheel is taken in from the driving torque sensor 5 so that the measured tire braking / driving force of each wheel is calculated.

  The measured tire braking / driving force calculated by the tire braking / driving force calculation unit 15 is transmitted to the tire braking / driving force filter unit 16. The tire braking / driving force filter unit 16 is constituted by a Kalman filter as will be described later, and in order to remove noise that increases due to differential processing of the wheel speed, an estimated tire braking force is estimated based on an estimated value using a tire model. The driving force is calculated.

  A wheel speed signal of each wheel is transmitted from the wheel speed sensor 3 to the tire braking / driving force filter unit 16, and the above-described measured tire braking / driving force is transmitted from the tire braking / driving force calculation unit 15. An estimated vehicle speed signal is transmitted from the vehicle speed estimation unit 13.

  Then, the estimated tire braking / driving force calculated (filtered) by the tire braking / driving force filter unit 16 is output to the vehicle attitude control device 17. Further, the road surface μ signal calculated by the tire model 18 used in the tire braking / driving force filter unit 16 is transmitted to the vehicle body speed estimation unit 13 described above.

Next, a control flow for vehicle speed estimation of the vehicle state estimation device will be described.
3 and 4 are control flows of the estimated vehicle speed estimation calculation method, and the estimated vehicle speed estimation method will be described according to this control flow.

First, in S1, the front-rear G signal a _mea and the wheel speed signals V _w1 , V _w2 , V _w3 , V _w4 are read from the front-rear G sensor 4 and each wheel speed sensor 3.

Next, in S2, the front-rear G signal a _mea is filtered by the acceleration filter unit 10 (a _mea → a _x ). In this filtering process, noise in the high frequency region of the front and rear G signal a _mea , which is raw data, is removed to make the signal easy to perform arithmetic processing.

In S3, by using a slip ratio map shown in FIG. 5, the longitudinal G signal a _x after the filtering process, calculates an equivalent slip ratio lambda _avg.
A specific calculation method will be described. First, the tire braking / driving force F _avg is obtained by the following equation.

すなわち、車体に生じる前後Ｇを、４つの車輪が発生するタイヤ制駆動力Ｆ１〜Ｆ４と比例すると擬制して、前後Ｇの値ａ_ｘからタイヤ制駆動力Ｆ_ａｖｇを演算して算出するのである。

That is, the longitudinal G generated in the vehicle body, four and fiction the wheel is proportional to tire longitudinal force F1~F4 generated is to calculate by calculating the tire longitudinal force F _avg from the value a _x of longitudinal G .

Since this tire braking / driving force F _avg can be considered as an average value of the tire braking / driving forces F1 to F4 of the four wheels, an average slip ratio of all the wheels, that is, an equivalent slip ratio is determined from the slip ratio map of FIG. Calculate λ _avg .

The slip ratio map has a proportional characteristic in which the slip ratio λ increases as the tire braking / driving force F increases up to a predetermined saturation slip ratio λ _{fel. avg} is calculated.

The slip ratio map is configured to change depending on the road surface μ. When the road surface μ is high, the map is changed to a map having a high tire braking / driving force. When the road surface μ is low, the map is changed to a map having a low tire braking / driving force. By doing so, it is possible to calculate the equivalent slip ratio λ _avg closer to the actually measured value. Further, FIG. 5 shows the slip ratio characteristic line only on the drive side, but the characteristic line is also symmetrical on the brake side.

Thus, after _obtaining the equivalent slip ratio λ _avg in S3, the process _proceeds to S4.
In S4, the reference vehicle speed _Vth is calculated by adding the previous estimated vehicle speed _{Vest (t-1)} and the differential value of the previous front and rear Gax _(t-1) . The reference vehicle speed _Vth is a vehicle speed that serves as a reference when calculating the wheel speed average of the four wheels. Specifically, it is calculated by the following formula.

次に、Ｓ５では、こうした求めた基準車体速Ｖ_ｔｈを基準にして、車輪速が所定値以上離れた場合、すなわち、車輪速が基準車体速Ｖ_ｔｈよりも所定値（例えば５ｋｍ／ｈ）以上大きい場合、又は車輪速が基準車体速Ｖ_ｔｈよりも所定値（例えば５ｋｍ／ｈ）以上小さい場合には、その車輪はスリップ輪として判断する。

Next, in S5, when the wheel speed is separated by a predetermined value or more with reference to the obtained reference vehicle body speed _Vth , that is, the wheel speed is a predetermined value (for example, 5 km / h) or more than the reference vehicle body speed _Vth. If it is larger, or if the wheel speed is smaller than the reference vehicle body speed _Vth by a predetermined value (for example, 5 km / h) or more, the wheel is determined as a slip wheel.

  In S6, it is determined whether or not the wheels determined as the slip wheels are all wheels. If it is determined that all the wheels are slip wheels (YES determination), the process proceeds to S9. If it is determined that all the wheels are not slip wheels (NO determination), the process proceeds to S7.

In S7, a wheel speed average V _wave is calculated for a non-slip wheel, that is, a non-slip wheel. Specifically, when there are no slip wheels, the wheel speed average is calculated by adding the wheel speeds of all the wheels (four wheels) and dividing by the total number of wheels. On the other hand, if there is even one slip wheel, the wheel speed average V _wave is calculated by adding the wheel speeds of the non-slip wheels and dividing the sum by the number of non-slip wheels.
Thus, by calculating the wheel speed average V _wave excluding slip wheels, it is possible to calculate the wheel speed average V _wave with the highest possible accuracy.

In S8, the estimated vehicle speed V _est is calculated from the wheel speed average V _wave and the equivalent slip ratio λ _avg by using the following equation.

この式は、加速時のスリップ率を算出する式と、減速時のスリップ率を算出する式を、それぞれ変形することで、推定車体速Ｖ_ｅｓｔを、等価スリップ率λ_ａｖｇと車輪速平均Ｖ_ｗａｖｅとから算出できるようにした式である。

This formula is obtained by modifying the formula for calculating the slip ratio during acceleration and the formula for calculating the slip ratio during deceleration, respectively, so that the estimated vehicle body speed V _est is _converted into the equivalent slip ratio λ _avg and the average wheel speed V _wave. This is an equation that can be calculated from

Thus, by calculating the estimated vehicle speed V _est, the estimated vehicle speed V _est of acceleration and deceleration can be calculated appropriately.

On the other hand, if YES is determined in S6, the process proceeds to S9 to calculate the reference vehicle speed _Vth as the estimated vehicle speed _Vest . That is, since all are judged as slip wheels, the wheel speed that can be the average of the wheel speeds cannot be taken in, so the reference vehicle body speed _Vth is substituted for the estimated vehicle body speed _Vest .

Thus, the estimated vehicle speed V _est calculated in S9 and S8 is output to the vehicle attitude control device 17 in S10. In this way, the estimated vehicle speed V _est calculated by the calculation process is output to the vehicle attitude control device 17, so that even if the vehicle speed is difficult to measure, the vehicle speed can be easily and accurately. Can be obtained.

Although not specifically shown, the estimated vehicle body speed V _est is set to be stored in the memory unit 12 every time it is calculated, and the next reference vehicle body speed V _th is calculated. It is used in cases.

Next, the control flow of tire braking / driving force estimation of this vehicle state estimation device will be described.
FIG. 6 is a control flow of an estimation calculation method of the estimated tire braking / driving force, and the estimation method of the estimated tire braking / driving force will be described according to this control flow.

First, in S11, from the wheel speed sensors 3 ... and the drive torque sensor 5 ..., the wheel speed signals _{V w1, V w2, V w3} , V w4 and the drive torque signal _{T 1, T 2, T 3} , T ₄ is read.

Next, in S <b> 12, each wheel speed signal obtained from each wheel speed sensor 3 is subjected to differentiation processing by the differentiation processing unit 14. That is, each wheel angular acceleration _dw1 , _dw2 , _dw3 , _dw4 is calculated from each wheel speed signal.

Then, in S13, the driving torque _{T 1, T 2, T 3} , T 4 obtained from the driving torque sensor 5, from the wheel angular acceleration _{d w1, d w2, d w3} , d w4 obtained by the differential processing unit 14, The measured tire braking / driving forces F _{x1 meas} , F _{x2 meas} , F _{x3 meas} , and F _{x4 meas} are calculated. Specifically, the measured tire braking / driving force is calculated using the following equation.

この式は、タイヤの運動方程式である以下の式を、変形することによって、構成している。

This equation is constructed by modifying the following equation, which is the equation of motion of the tire.

前述の計測タイヤ制駆動力を求める式のうち、タイヤ慣性モーメントＪと、タイヤ半径ｒは、車両諸元であり、一定値であるため、変数である駆動トルクＴと、車輪角加速度ｄwの数値を代入することにより、計測タイヤ制駆動力を算出することになる。

Of the above-described equations for calculating the measured tire braking / driving force, the tire inertia moment J and the tire radius r are vehicle specifications and are constant values. Therefore, the driving torque T and the numerical values of the wheel angular acceleration dw are variables. By substituting, the measured tire braking / driving force is calculated.

Next, in S14, the measured tire braking / driving forces Fx1 _meas , Fx2 _meas , Fx3 _meas , Fx4 _meas are taken into the Kalman filter (16), and compared with the so-called tire model 18, the estimated tire braking / driving forces Fx1 _est , Fx2 _est , Fx3 _est , Fx4 _est are calculated.

Here, the estimated tire braking / driving force F _xest is calculated using the Kalman filter (16) because the differential processing is performed when the wheel angular acceleration dw is calculated. Since the calculated noise disturbs the calculated value and a control delay occurs due to the time constant, the estimated tire braking / driving force is calculated using the Kalman filter (16) in order to eliminate the noise and the control delay.

  Here, the Kalman filter will be briefly described. The Kalman filter is designed in advance by designing a predetermined virtual model that incorporates information such as signal noise and measurement error, and by changing this virtual model according to the input signal, the most probable value in statistical probability theory. Are output as estimated values.

For this reason, by using a Kalman filter in the tire braking / driving force filter unit 16, the measured tire braking / driving forces _Fx1meas , _Fx2meas , _Fx3meas , and _{Fx4meas in} which noise and control delay are generated due to differential processing are _reduced to noise and control delay. It can be calculated as the estimated tire braking / driving force F _x1est , F _x2est , F _x3est , and F _x4est with suppressed.

Next, in S _<b> 15, the calculated estimated tire braking / driving forces F _x1est , F _x2est , F _x3est , F _x4est are output to the vehicle attitude control device 17. Since the estimated tire braking / driving force can be obtained on the vehicle attitude control device 17 side, a change in the vehicle behavior can be predicted in advance, so that it is possible to perform a control that further increases the safety of the vehicle behavior.

  In S16, the road surface μ is estimated and calculated from the tire model 18 and the tire braking / driving force.

  In S17, the next slip ratio map is changed according to the road surface μ estimated and calculated in S16. That is, since the estimated road surface μ changes, the slip ratio map is changed so that the estimated value is closer to the actual value.

  Through the above steps, one control flow of the vehicle state estimation device is completed, and thereafter, the next control flow is repeatedly performed.

Next, the effect of this embodiment comprised in this way is demonstrated.
The vehicle state estimation device of this embodiment includes a front / rear G sensor 4 for detecting the longitudinal acceleration of the vehicle body and a wheel speed sensor 3 for detecting the wheel speed of each wheel 2. The equivalent slip ratio λ _avg is calculated, and the slip state of the wheel is determined based on the reference vehicle body speed V _th and the wheel speeds V _w1 , V _w2 , V _w3 , V _{w4 of} each wheel 2. The wheel speed average V _wave is calculated from the non-slip wheel, and the estimated vehicle speed V _est is calculated from the equivalent slip ratio λ _avg and the wheel speed average V _wave .

Thus, the estimated vehicle body speed V _est can be calculated using only the generally used wheel speed sensor 2 and front / rear G sensor 4 without using an expensive sensor such as a yaw rate sensor or a lateral acceleration sensor.
In addition, since the wheel speed average is calculated for the non-slip wheels except for the slipping wheel, the wheel speed average V _wave can be calculated without taking in a wheel speed that is significantly different from the vehicle body speed. For this reason, the calculation accuracy of the estimated vehicle body speed V _est can be increased.
Therefore, in the vehicle state estimation device for estimating the vehicle state quantity in the four-wheel drive vehicle, the vehicle state quantity is estimated by detecting the vehicle state by using as few sensors as possible, and thus the vehicle state quantity can be simplified. In addition, it can be estimated with high accuracy.

Further, in this embodiment, the reference vehicle body speed _Vth is calculated by adding the previous estimated vehicle body speed _{Vest (t-1)} and the differential value of the previous front and rear Gax _(t-1) .
As a result, the reference vehicle body speed _Vth for determining the slip state of the wheel can be obtained without providing a separate sensor or the like for detecting a reference vehicle body speed with poor feasibility.
Therefore, it is possible to obtain the reference vehicle speed _Vth without providing a sensor or the like that detects a reference vehicle speed with poor feasibility.

Further, in this embodiment, when all the wheels slip to a predetermined value or more, the reference vehicle body speed _Vth is set to be calculated as the estimated vehicle body speed _Vest .
Thereby, when all the wheels slip to a predetermined value or more and there is no non-slip wheel, the wheel speed average V _wave cannot be calculated by directly calculating the reference vehicle speed V _th as the estimated vehicle speed V _est. Even in this case, the estimated vehicle speed V _est can be calculated.
Therefore, the estimated vehicle body speed V _est can be calculated even when the wheel speeds detected by the all-wheel slip are all greatly different from the vehicle body speed.

In this embodiment, the slip ratio map is changed according to the road surface μ.
Thereby, since the calculated equivalent slip ratio changes according to the road surface μ, the equivalent slip ratio λ _avg closer to the actual value can be calculated.
Therefore, the estimated vehicle speed V _est can be calculated with a value closer to the actually measured value, reflecting the difference in the road surface μ.

Further, in this embodiment, a wheel torque acceleration d _w1 , d _w2 , d _w3 , d calculated by differentiating the wheel speed by the differentiation processing unit 14 is provided, which includes a driving torque sensor 5 that detects a driving torque for driving the wheel. _The tire braking / driving force (measured tire braking / driving) is determined by _w4 , the estimated vehicle speed V _est calculated by the vehicle speed estimator 8 and the driving torques T ₁ , T ₂ , T ₃ , T ₄ detected by the driving torque sensor 5. Forces F _{x1 meas} , F _{x2 meas} , F _{x3 meas} , F _{x4 meas} ).
Thus, the braking / driving force of the tire can be calculated from the driving torque detected by the driving torque sensor 5, the estimated vehicle body speed, and the angular acceleration of the wheel.
For this reason, it is possible to calculate the tire braking / driving force generated by each tire, which has been difficult to detect in the past, and the vehicle attitude control device can be controlled using the calculated tire braking / driving force. This can be done after predicting the behavior of the vehicle.
Therefore, it is possible to configure a vehicle attitude control device that calculates in advance the tire braking / driving force, which has been difficult to detect, and further improves safety performance and the like.

In this embodiment, the tire braking / driving force filter unit 16 includes the estimated vehicle body speed V _est , the measured tire braking / driving forces F _{x1 meas} , F _{x2 meas} , F _{x3 meas} , F _{x4 meas} and the wheel speeds V _w1 , V _w2 , V _w3 , _Vw4 is taken in and compared with the tire model 18 to form an estimated value Kalman filter.
Thereby, the value of the tire braking / driving force (estimated tire braking / driving force F _x1est , F _x2est , F _x3est , F _x4est ) is calculated by the estimated value of the Kalman filter 18.
Therefore, a calculated value measured tire longitudinal force _{F x1meas, F x2meas, F x3meas} , can be suppressed as much as possible the occurrence of noise _{F X4meas,} it is possible to prevent the occurrence of control delay.
Therefore, the value of the tire braking / driving force can be calculated more accurately and quickly. By using the estimated tire braking / driving forces F _x1est , F _x2est , F _x3est , F _x4est , the performance of the vehicle attitude control device can be improved. Can be increased.

In addition, what removes this noise and control delay is not necessarily limited to a Kalman filter as in this embodiment. Further, the tire model may also be designed by taking in information on a tire contact load, information on secular change, information on tire pressure, and the like.
Further, the above-described slip ratio map is not limited to this, and a “table”, an “expression”, or the like may be substituted.

As described above, in the correspondence between the configuration of the present invention and the above-described embodiment,
The longitudinal acceleration detecting means of the present invention corresponds to the longitudinal G sensor 4 of the embodiment,
Similarly,
The wheel speed detection means corresponds to the wheel speed sensor 3,
The equivalent slip ratio calculating means corresponds to the vehicle body speed estimating unit 13,
The wheel speed average calculating means corresponds to the wheel speed filter unit 11,
The estimated vehicle speed calculation means corresponds to the vehicle speed estimation unit 13,
The tire drive torque detection means corresponds to the drive torque sensor 5,
The tire braking / driving force estimation means corresponds to the tire braking / driving force estimator 9,
The present invention is not limited to the above-described embodiment, but includes an embodiment applied to any vehicle state estimation device.

  In particular, the vehicle drive system is not limited to the four-wheel independent electric motor system drive system as in the above-described embodiment, but is a general drive system composed of an engine and a transmission. Also good. The detection of the drive torque in this case may be performed by multiplying the engine output by the transmission gear ratio.

The schematic schematic diagram of the motor vehicle which employ | adopted the vehicle state estimation apparatus which concerns on embodiment. The system block diagram of a vehicle state estimation apparatus. The first half part of the flowchart explaining the control method of the vehicle body speed estimation of the vehicle state estimation device. The latter half part of the flowchart explaining the control method of vehicle speed estimation. A map for obtaining the equivalent slip ratio. The flowchart explaining the control method of tire braking / driving force estimation.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Automobile 2 ... Wheel 3 ... Wheel speed sensor 4 ... Front-back G sensor 5 ... Drive torque sensor 6 ... Electric motor 7 ... Control unit 8 ... Vehicle body speed estimator 9 ... Tire braking / driving force estimator 13 ... Vehicle body speed estimation part 15 ... Tire braking / driving force calculation unit 16 ... Tire braking / driving force filter unit

Claims (6)

A vehicle state estimation device for estimating a state amount of a vehicle in an all-wheel drive vehicle,
Longitudinal acceleration detection means for detecting longitudinal acceleration of the vehicle body;
Wheel speed detection means for detecting the wheel speed of each wheel;
Equivalent slip ratio calculating means for calculating an equivalent slip ratio of the slip state of all wheels from the detected longitudinal acceleration;
A wheel speed average calculating means for determining a slip state of each wheel based on a reference vehicle body speed and a wheel speed of each wheel, and calculating a wheel speed average from a non-slip wheel having a slip state of a predetermined value or less;
A vehicle state estimation device comprising: estimated vehicle body speed calculating means for calculating an estimated vehicle body speed from the equivalent slip ratio and the average wheel speed. The vehicle state estimation device according to claim 1, wherein the reference vehicle body speed is calculated by adding a differential value between the previous estimated vehicle body speed and the previous longitudinal acceleration. The vehicle state estimation device according to claim 1 or 2, wherein when all the wheels slip to a predetermined value or more, the reference vehicle body speed is set to be calculated as an estimated vehicle body speed. The vehicle state estimation device according to any one of claims 1 to 3, wherein a calculated value of the equivalent slip ratio calculating means is changed according to a road surface μ. Tire driving torque detecting means for detecting driving torque for driving each wheel;
Wheel angular acceleration calculation means for differentiating the wheel speed detected by the wheel speed detection means to calculate the angular acceleration of each wheel;
The braking / driving force of each tire is estimated from the estimated vehicle speed calculated by the estimated vehicle speed calculating means, the tire driving torque detected by the tire driving torque detecting means, and the wheel angular acceleration calculated by the wheel angular acceleration calculating means. The vehicle state estimation device according to claim 1, further comprising: a tire braking / driving force estimation unit that performs The tire braking / driving force estimating means includes a Kalman filter that calculates an estimated value by taking at least an estimated vehicle body speed, a tire driving torque, and a wheel angular acceleration and comparing the tire braking / driving force estimating means with a predetermined tire model. The vehicle state estimation device according to claim 5.

Images