JPH0834357A - Controller of vehicle - Google Patents

Abstract

PURPOSE:To improve control reliability and secure safety by preventing control that is based on an erroneous yaw rate surmise. CONSTITUTION:The rotational speeds Wfl, Wfr, Wrl, Wrr of right and left front wheels and rear wheels are detected by means of four vehicle wheel speed sensors 31-34. The rotational speed difference Rv of right and left rear wheels that are follower wheels is calculated by means of a calculating means 41, and a yaw rate phi is surmised by means of a neural net operation portion 44 that makes the rotary speed difference Rv or the like an input variable. On the basis of a surmised yaw rate phi, the steering control amount Rs of the rear wheels is set. The steering control amount Rs is decided, for example, by adding a vehicle speed induction control term that makes vehicle speed V a factor and a yaw rate compensation term that makes the yaw rate phia factor, and a reliability deciding means 52 that decides the reliability of the surmised yaw rate phi on the basis of the rotary speeds Wfl-Wrr of the four wheels, and a regulating means 53 that makes the yaw rate compensation term in the calculation expression of the steering control amount Rs a zero when there is no reliability of the surmised yaw rate phi, are provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device mounted on a vehicle, and more particularly, in a four-wheel steering device or the like, a control target (for example, a four-wheel steering device) is controlled based on a yaw rate estimated from a rotational speed difference between left and right wheels. Rear wheel) in the control.

[0002]

2. Description of the Related Art Conventionally, a vehicle may be equipped with a four-wheel steering device that steers the rear wheels when steering the front wheels. As a four-wheel steering system of this type, for example, Japanese Patent Laid-Open No.
As disclosed in Japanese Patent No. 185947, there is known a vehicle speed and yaw rate detected by sensors, and steering of rear wheels is controlled based on these detected values.

A conventional yaw rate detecting or estimating device (generally referred to as a yaw rate sensor) detects the rotation speeds of the left and right wheels by the wheel speed sensors, respectively, and the difference in the rotation speeds of the left and right wheels is calculated from the detected values. Is generally calculated, and the yaw rate is estimated in a linear form derived from the geometrical relationship (so-called two-wheel model) at the time of circular turning based on this rotational speed difference (for example, see Japanese Patent Laid-Open No. 2-40504). ).

Further, in the above-mentioned conventional yaw rate sensor,
Since the yaw rate is estimated in a linear form derived from the geometrical relationship during circular turning, there is a drawback that the yaw rate estimation accuracy is poor when the motion state is in a nonlinear region. Those using a network have been proposed. This neural network is usually the difference in rotational speed of the left and right wheels, which is related to yaw rate,
The vehicle state quantities such as the steering angle, the steering speed, the vehicle speed, and the vehicle body acceleration are used as input variables to synapse-connect with the neurons of the intermediate layer, and the neurons of the intermediate layer are synapse-connected with the neurons of the output layer that outputs the estimated yaw rate value. It has been demonstrated that the yaw rate can be estimated with high accuracy from the linear region to the non-linear region by increasing the number of learnings.

[0005]

By the way, the yaw rate is estimated from the rotational speed difference between the left and right wheels, including the one using a neural network, and the controlled object is controlled based on the estimated yaw rate, the left and right wheels are rotated. The calculation of the speed difference is very important for control, but conventionally, the rotational speed of the left and right wheels is detected by only one of the front and rear wheels, which are single driven wheels. It just calculates the difference. For this reason, for example, if one of the left and right driven wheels slips on the road surface or becomes in a brake locked state, the yaw rate is erroneously estimated from the difference in rotational speed between the left and right wheels at this time, and erroneous control may occur. There is a problem of being a cause.

The present invention has been made in view of the above points, and an object of the present invention is to detect not only the driven wheels but also the rotation speeds of all four wheels, and the rotation of the left and right wheels from these detected values. By calculating the speed difference and then estimating the yaw rate, the reliability of the yaw rate estimation is judged at the same time, and the control based on the incorrect yaw rate estimation is prevented, so that the safety can be secured and the control reliability can be improved. It is intended to provide a vehicle control device.

[0007]

In order to achieve the above object, the invention according to claim 1 is a vehicle control device for controlling a controlled object based on a yaw rate estimated from a rotational speed difference between the left and right wheels. The four wheel speed detecting means for respectively detecting the rotation speeds of the front wheels and the rear wheels, the calculating means for obtaining the rotation speed difference between the left and right wheels from the rotation speeds of the left and right front wheels or the rotation speeds of the left and right rear wheels, and the above calculating means. Based on the yaw rate estimating means for estimating the yaw rate from the obtained rotational speed difference between the left and right wheels, and the rotational speed of the wheel and the rotational speed of the other wheels used for calculating the left and right wheel rotational speed difference by the calculating means Reliability judging means for judging the reliability of the yaw rate estimated by the means, and regulation for restricting control based on the estimated yaw rate when the estimated yaw rate is not reliable as a result of the judgment by the judging means. A configuration and a stage.

In the invention according to claim 2, in the invention according to claim 1, when the control target is the rear wheel steered at the time of steering the front wheels, that is, the vehicle control device is applied to a four-wheel steering device. Indicate the case.

The inventions according to claims 3 and 4 both show one aspect of the invention according to claim 2. That is, in the invention according to claim 3, the control amount of the rear wheel steering based on the yaw rate is determined by adding the vehicle speed sensitive control term having the vehicle speed as a factor and the yaw rate compensation term having the yaw rate as a factor. Yes, the regulation means makes the yaw rate compensation term zero when the estimated yaw rate is not reliable. In addition, claim 4
In the invention according to claim 1, the control of the rear wheel steering based on the yaw rate is a yaw rate feedback control for controlling the actual yaw rate so as to match the target yaw rate at which the sideslip angle is zero. When there is not, the yaw rate feedback control is stopped, and the steering of the rear wheels is controlled based on the map set in advance with the vehicle speed as a factor.

The invention according to claim 5 shows one mode of the reliability judgment by the reliability judgment means which is a component of the invention according to claim 1. That is, the reliability determining means obtains a rotational speed difference between the left front wheel and the left rear wheel and a rotational speed difference between the right front wheel and the right rear wheel. When the rotational speed difference is smaller than a predetermined value, the yaw rate reliability is reduced. It is determined that the yaw rate is not reliable when at least one of the two rotational speed differences is larger than a predetermined value.

[0011]

With the above construction, in the invention according to claim 1, the rotational speeds of the four wheels, that is, the left and right front wheels and the rear wheels are respectively detected by the wheel speed detecting means, and in the calculating means for receiving these detection signals, for example, The yaw rate estimating means estimates the yaw rate from the rotational speed difference between the left and right wheels from the rotational speeds of the left and right front wheels, which are driving wheels, or the rotational speeds of the left and right rear wheels, and the yaw rate is estimated based on the estimated yaw rate. (In the invention according to claim 2, the rear wheels in the four-wheel steering system) are controlled. At the same time as the calculation of the rotational speed difference between the left and right wheels or the estimation of the yaw rate, the reliability determination means makes the above estimation based on the rotational speed of the wheel used for calculating the rotational speed difference between the left and right wheels and the rotational speed of the other wheels. The reliability of the yaw rate is judged. For example, in the invention according to claim 5, it is determined that the yaw rate is reliable when the rotational speed difference between the left front wheel and the left rear wheel and the rotational speed difference between the right front wheel and the right rear wheel are smaller than predetermined values, respectively. When at least one of the two rotation speed differences is larger than a predetermined value, it is determined that the yaw rate is not reliable.

When the reliability of the estimated yaw rate is unreliable as a result of the determination by the reliability determination means, the regulation means regulates the control based on the estimated yaw rate. For example, in the invention according to claim 3, when the control amount of the rear wheel steering based on the yaw rate is determined by adding the vehicle speed sensitive control term and the yaw rate compensation term, the yaw rate compensation term is set to zero. Further, in the invention according to claim 4, when the control of the rear wheel steering based on the yaw rate is the yaw rate feedback control, the yaw rate feedback control is stopped and the rear wheels are controlled by the map control. As a result, control based on incorrect yaw rate estimation is prevented, and safety and the like are secured.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a schematic construction of a four-wheel steering system for a vehicle according to a first embodiment of the present invention. 1 is a steering wheel, 2L and 2R are left and right front wheels, 3L and 3R are left and right rear wheels, 10 Is a front wheel steering device for steering the left and right front wheels 2L, 2R by operating the steering wheel 1, 20
Is a rear wheel steering device that steers the left and right rear wheels 3L and 3R in response to steering of the front wheels 2L and 2R by the front wheel steering device 10. The vehicle of this embodiment is an FF vehicle (not shown), in which the front wheels 2L and 2R serve as driving wheels, and the rear wheels 3L and 3R.
Is the driven wheel.

The front wheel steering device 10 has a relay rod 11 arranged in the vehicle width direction, and both ends of the rod 11 are tie rods 12, 12 and a knuckle arm 13, respectively.
It is connected to the left and right front wheels 2L, 2R via 13.
The relay rod 11 is provided with a rack and pinion mechanism 14 that moves the relay rod 11 left and right in conjunction with the operation of the steering wheel 1, and when the steering wheel 1 is operated, the rack and pinion mechanism 14 is moved by an angle corresponding to the operation amount. It is configured to steer the front wheels 2L, 2R.

On the other hand, the rear wheel steering device 20 has a relay rod 21 arranged in the vehicle width direction like the front wheel steering device 10, and both ends of the rod 21 are tie rods 22 and 22 and a knuckle arm, respectively. The left and right rear wheels 3L, 3R are connected via 23, 23. A centering spring 24 for urging the rod 21 to a neutral position is arranged on the relay rod 21, and a rack and pinion mechanism 25 is arranged on the relay rod 21.
6, the speed reduction mechanism 27 and the motor 28 are linked, and the relay rod 21 is moved in the vehicle width direction via the rack and pinion mechanism 25 by the rotational driving of the motor 28 when the clutch 26 is engaged, and the rear wheels 3L, 3R. Is configured to be steered by an angle according to the rotation amount of the motor 28.

The operation of the motor 28 is controlled by the control unit 30. 31, 32, 33 and 34
Are wheel speed sensors as four wheel speed detecting means for detecting rotational speeds (hereinafter, referred to as wheel speeds) of the left front wheel 2L, the right front wheel 2R, the left rear wheel 3L and the right rear wheel 3R, respectively, and 35 is a steering wheel 1. The steering angle sensor detects a steering angle, and signals from these sensors 31 to 35 are all input to the control unit 30.

In the control unit 30, as shown in FIG. 2, the wheel speed Wr of the left and right rear wheels 3L and 3R which are the driven wheels detected by the wheel speed sensors 33 and 34.
The rotational speed difference Rv between the left and right wheels is calculated by subtracting l and Wrr from each other at the addition point 41. The addition point 41 functions as a calculating means for obtaining the rotational speed difference Rv between the left and right wheels. Have. The rotational speed difference Rv is input to the neural net calculation unit 44 through the filter 42 and the normalization processing unit 43.

Further, in the control unit 30, the wheel speeds Wrl and Wrr of the left and right rear wheels 3L and 3R are added to each other at the addition point 45, and this addition value is halved by the coefficient integration section 46. Thus, the vehicle speed V is calculated, and the wheel speed sensors 33 and 34, the addition point 45, and the coefficient integrating unit 46 constitute a vehicle speed detecting means 47 that detects the vehicle speed V. The vehicle speed V is input to the neural network calculation unit 44 through the normalization processing unit 43. The steering angle Fs detected by the steering angle sensor 35 is also input to the neural net calculation unit 44 through the normalization processing unit 43. Further, the steering angle Fs is
The steering speed Dfs is calculated by inputting as it is and after being delayed by one measurement timing in the delay circuit 49, and subtracting both input values at the addition point 48 from each other. In addition to the sensor 35, the addition point 48 and the delay circuit 49 constitute a steering speed detecting means 50 for detecting the steering speed Dfs. The steering speed Dfs is also input to the neural network calculation unit 44 through the normalization processing unit 43.

The neural network computing unit 44 normalizes the rotational speed difference of the left and right wheels Rvn, the vehicle speed normalized value Vn, the steering angle normalized value Fsn, and the steering speed normalized that are respectively normalized by the normalization processing unit 43. The neural network has a value Dfsn as an input variable and has a plurality of neurons (six in the case of this embodiment) in the intermediate layer. The yaw rate estimating means calculates the estimated yaw rate φ by a neural network operation. This estimated yaw rate φ is input to the control amount setting means 51. The control amount setting means 51
Receives the vehicle speed V detected by the vehicle speed detecting means 47 and the steering angle Fs detected by the steering angle sensor 35 together with the estimated yaw rate φ from the neural network calculation unit 44, and based on these, the rear wheels 3L, 3R. The steering control amount Rs is set and output to the motor 28.

Further, the control unit 30 includes
A reliability determination means 52 and a regulation means 53 for receiving wheel speed signals from the four wheel speed sensors 31 to 34 are provided. The reliability determining means 52 uses four wheel speeds Wfl,
Based on Wfr, Wrl, and Wrr, the reliability of the yaw rate φ estimated by the neural network operation unit 44 is judged, and the regulation means 53 is judged by the judging means 52. When it is determined that there is not, the estimated yaw rate φ in the control amount setting means 51
The setting of the control amount Rs based on the above is regulated.

Next, the operation control of the motor 28 by the control unit 30 and the steering control of the rear wheels 3L, 3R will be described with reference to the main flow chart shown in FIG.

In FIG. 3, first, after waiting for the measurement timing in step S1, the wheel speeds Wfr, Wfl, Wrr, Wrl of the four wheels as the vehicle motion information in step S2.
Are measured by the wheel speed sensors 31 to 34, respectively, and the steering angle Fs as driver operation information is measured by the steering angle sensor 35 in step S3. Subsequently, in step S4, the wheel speed W of the left and right rear wheels which are the driven wheels is set.
The rotational speed difference Rv between the left and right wheels, which is the difference between rl and Wrr, is obtained at the addition point 41, and the filter calculation is performed by the filter 42 on the rotational speed difference Rv. This rotation speed difference Rv
The filter calculation of Rvf = Filter (Wrl-Wrr) = b (1) -Rv (t) + b (2) -Rv (t-1) + b (3) -Rv (t-3) + b ( 4) ・ Rv (t-4) + b (5) ・ Rv (t-5) -a (2) ・ Rvf (t-1) -a (3) ・ Rvf (t-2) -a (4) ・Rvf (t-4) -a (5) -Rvf (t-5) ... However, the filter constant is, for example, a = [1.0-2.5562 2.6311-1.2487 0.2299] b = [0.0199-0.0052 0.0266-0.0052 0.0199].

After the above filter calculation, in step S5, the normalization processing unit 43 normalizes the rotational speed difference Rv of the left and right wheels, the vehicle speed V, the steering angle Fs, and the steering speed Dfs, which are the input information. This normalization calculation is performed by the following equation: Rvn = Rv / 2 Vn = (V-70) / 70 Fsn = Fs / 5 Dfsn = (Fs (t) -Fs (t-1)) / 20 P = [Vn, Rvn, Fsn, Dfsn] ^T. The T of the matrix P means a transposed matrix.

Then, in step S6, the neural network computing unit 44 normalizes the rotational speed difference Rvn, the vehicle speed normalized value Vn, the steering angle normalized value Fsn, and the steering speed normalized value Dfsn.
An estimated yaw rate φ is calculated by performing a neural network operation using as an input variable. The calculation of the estimated yaw rate φ (calculation of the neural network) is performed according to the flowchart shown in FIG.

Then, in step S7, the estimated yaw rate φ is used to determine the yaw rate sensitive control amount ry by the following equation: ry = f1 (V) × f2 (Fs) × φ.

However, f1 (V) is a functional expression of the vehicle speed response coefficient Kv having the vehicle speed V as a factor. For example, as shown in FIG.
Is set to a large value and the coefficient Kv is set to be small because the vehicle turns too sharply at high vehicle speed. Further, f2 (Fs) is a functional expression of the steering angle response coefficient Kf having the steering angle Fs as a factor. For example, as shown in FIG. 5, when the steering angle Fs is small, the stability near the neutral is emphasized and the coefficient Kf. Is set to be small, and when the steering angle Fs is large, the coefficient Kf is set to be large in order to improve the effectiveness of the steering wheel at the time of sudden turning of the large steering angle.

Then, in step S8, the coefficient Kv is set by using the preset map. In the above map, the coefficient Kv is set to increase gradually as the vehicle speed V increases, and the increase amount decreases gradually.

Then, in step S9, the estimated yaw rate φ
Evaluate the reliability of. This reliability evaluation is performed according to the flowchart shown in FIG. 8. When there is reliability, the reliability flag Fa is set, and when there is no reliability, the reliability flag Fa is reset. And step S
At 10, it is determined whether the reliability flag Fa is 1, that is, whether the estimated yaw rate φ is reliable, and if so, at step S11, the steering control amount Rs of the rear wheels 3L, 3R is determined.
Is set by the following formula Rs = Kv × Fs + ry ... On the other hand, when there is no reliability, step S
At 12, the steering control amount Rs of the rear wheels 3L and 3R is set by the following formula Rs = Kv × Fs.

Thereafter, the steering control amount Rs is set to the motor 2
Output to S8 and return to step S1.

Here, the first term on the right side of the above equation is a vehicle speed sensitive control term which is the product of the coefficient Kv having the vehicle speed V as a factor and the steering angle Fs, and the second term is obtained from the above equation. It is a yaw rate compensation term. Also, the right side of the formula is
It consists only of the vehicle speed sensitive control term, and corresponds to the yaw rate compensation term of which is zero on the right side of the equation. Step S7,
The control amount setting means 51 executes S8, S11, S12, and the restricting means 53 executes steps S9, S10. Therefore, the control amount setting means 51 is adapted to set the steering control amount Rs of the rear wheels 3L, 3R by the arithmetic expression (equation) consisting of the vehicle speed sensitive control term and the yaw rate compensation term, and the regulating means 53 is When the estimated yaw rate φ is not reliable, the yaw rate compensation term in the above equation is set to zero.

Next, regarding the calculation of the neural network,
Description will be given according to the flowchart shown in FIG.

In FIG. 6, first, in step S31, the normalization information P (vehicle speed normalization value Vn, rotational speed difference normalization value Rv).
n, steering angle normalized value Fsn and steering speed normalized value Dfsn)
After inputting, the matrix product U0 of the normalized information P and the intermediate layer weighting coefficient W1 (i, j) is obtained in step S32, and the intermediate layer bias coefficient B1 (i) is added to this matrix U0 in step S33. , And replace the value with the matrix U0. Here, i is the number of neurons in the intermediate layer and j is the number of inputs. In the case of this embodiment, i = 6 and j =
It is 4.

Then, in step S34, the intermediate layer transfer function U (= ta), which is the tangent hyperbolic function of the matrix U0.
nh (U0)) is calculated. The use of the tangent hyperbolic function as the intermediate layer transfer function corresponds to the fact that the input is distributed in both positive and negative. The calculation of this intermediate layer transfer function U is
This is performed according to the flowchart shown in FIG. Then, in step S35, the intermediate layer transfer function U and the output layer weighting coefficient W are obtained.
An estimated yaw rate φ, which is a matrix product with 2 (i), is obtained, and the output layer bias coefficient B2 is added to this estimated yaw rate φ in step S36, and the value is newly replaced with the estimated yaw rate φ. After that, the process returns to the main flowchart.

The intermediate layer weighting coefficient W1, the intermediate layer bias coefficient B1, the output layer weighting coefficient W2, and the output layer bias coefficient B2 are manufactured by learning using the back propagation method to manufacture the yaw rate estimating apparatus according to this embodiment. It was previously determined for all devices.

Next, regarding the calculation of the intermediate layer transfer function U,
This will be described with reference to the flowchart shown in FIG.

In FIG. 7, first, the variable ii is set to 1 in step S41, and then it is determined in step S42 whether the matrix value U0 (ii) is positive or negative.
If it is positive, the flag f is reset at step S43, and the process proceeds to step S46, while if it is negative, the flag f is set at step S44, and at step S45, the matrix value U0 (ii) is multiplied by a minus sign to make it positive. After replacing with the value of, the process proceeds to step S46. Step S
In 46, the above value U0 (i
The intermediate layer transfer function value U (ii) corresponding to i) is calculated. In the above map, U = tanh (U0) where 0≤U0≤
It is expressed in the range of 20. When the intermediate layer transfer function value U (ii) is calculated using the map in this way, the calculation time can be saved.

Thereafter, in step S47, it is determined whether or not the flag f is set. If the flag f is set, that is, if the matrix value U0 (ii) is set to a positive value, then this is determined in step S48. Multiply the value by a minus sign to restore the original negative value. Then, after incrementing the variable ii by 1 in step S49, the variable i is incremented in step S50.
It is determined whether i is larger than the order i of the input information of the neural network. When the determination is NO, the process returns to step S42, while when the determination is YES, the calculation of the intermediate layer transfer function is ended and the process returns to the neural network calculation.

Next, the reliability evaluation of the estimated yaw rate φ will be described with reference to the flowchart shown in FIG.

In FIG. 8, first, in step S51, the rotational speed difference between the front and rear wheels is calculated separately for the left and right wheels. That is, the rotational speed difference Dl (= Wfl-Wrl) between the left and right front wheels is calculated from the wheel speed Wfl of the left front wheel and the wheel speed Wrl of the left rear wheel, and
From the wheel speed Wfr of the right front wheel and the wheel speed Wrr of the right rear wheel, the rotational speed difference Dr (= Wfr-Wrr) between the front and rear right wheels is calculated. Then, in step S52, it is determined whether or not the absolute value of the rotational speed difference Dl of the left and right front wheels is larger than a predetermined value, and in step S53, whether or not the absolute value of the rotational speed difference Dr of the right and front wheels is larger than a predetermined value. Judgment, when both judgments are ON,
That is, when the rotational speed difference Dl, Dr between the front and rear wheels is smaller than a predetermined value for both the left and right wheels, the reliability flag Fa is set, while when either of the judgments is YES, that is, the rotational speed difference Dl, Dr between the front and rear wheels is either. When is larger than a predetermined value, the reliability flag Fa is reset.

Next, to explain the operation and effect of the first embodiment, the control unit 30 of the rear wheel steering system 20 will be described.
When setting the steering control amount Rs of the rear wheels, first, the wheel speeds Wfl, Wfr, Wrl, Wrr of the four wheels are detected by the wheel speed sensors 31 to 34, respectively, and among these wheel speeds, the driven wheel is the driven wheel. Based on the wheel speeds Wrl and Wrr of the rear wheels 3L and 3R, the rotational speed difference Rv between the left and right wheels is calculated at the addition point 41 as the calculating means, and the vehicle speed V is calculated at the vehicle speed detecting means 47. Then, the normalization processing unit 4 determines the rotational speed difference Rv between the left and right wheels and the vehicle speed V together with the steering angle Fs of the front wheels detected by the steering angle sensor 35 and the steering speed Dfs which is a differential value thereof.
After being normalized by 3, the neural network calculation unit uses these normalized values (rotational speed difference normalized value Rvn, vehicle speed normalized value Vn, steering angle normalized value Fsn and steering speed normalized value Dfsn) as input variables. At 44, a neural network operation is performed to calculate an estimated yaw rate φ. In this neural network calculation, the yaw rate φ can be estimated with high accuracy from the linear region to the non-linear region.

Then, the control amount setting means 51 sets the rear wheel steering control amount Rs on the basis of the estimated yaw rate φ, the vehicle speed V and the steering angle Fs and outputs it to the motor 28, whereby the rear wheels 3L, 3R. Steering is controlled. The steering control amount Rs is usually a vehicle speed sensitive control term (Kv × Fs
) Is added to the yaw rate compensation term (ry), the rear wheel 3L,
It is possible to appropriately steer the 3R.

At the same time when the yaw rate φ is estimated,
The wheel speeds Wfl, Wfr, Wr of the four wheels are determined by the reliability determining means 52.
Rotational speed difference between front and rear wheels Dl and Dr separately from l and Wrr
Is calculated, and the reliability of the estimated yaw rate φ is judged from the rotational speed difference Dl, Dr between the front and rear wheels. That is, when the two rotational speed differences Dl and Dr are smaller than the predetermined value Dl, it is determined that the yaw rate is reliable, and at least one of the two rotational speed differences Dl and Dr is the predetermined value Dl.
If it is larger than l, it is judged that the yaw rate is not reliable. When the estimated yaw rate φ is reliable, the steering control amount Rs is obtained by adding the yaw rate compensation term to the vehicle speed response control term as in the above-described normal control, but it is determined that the estimated yaw rate φ is not reliable. Then, the yaw rate compensation term is made zero by the restricting means 53, and the steering control amount Rs is calculated only by the vehicle speed response control term. As a result, it is possible to prevent the control based on the erroneous estimated yaw rate φ and improve the safety.

In the first embodiment, the reliability determining means 52 uses the four wheel speeds Wfl, Wfr, Wrl, Wrr.
In determining the reliability of the estimated yaw rate φ on the basis of the above, the rotational speed differences Dl, Dr of the front and rear wheels are calculated separately for the left and right wheels, and the rotational speed differences Dl, Dr of the front and rear wheels are respectively larger or smaller than a predetermined value. Although the reliability of the estimated yaw rate φ is determined depending on whether the reliability of the estimated yaw rate φ is determined according to the flowchart shown in FIG. 9 or the reliability of the estimated yaw rate φ is determined according to the flowchart shown in FIG. You may judge sex.

In FIG. 9, first, at step S51 ', the rotational speed difference Df (= Wfr-Wf) between the left and right wheels is separately set for the front and rear wheels.
l) and Dr (= Wrr-Wrl) are calculated. Succeedingly, in a step S52 ', it is determined whether or not an absolute value of the front wheel-side left / right wheel rotation speed difference Df is larger than a predetermined value Lm, and in a step S53', an absolute value of the rear wheel-side left / right wheel rotation difference Dr is a predetermined value Lm. If both of these judgments are ON, that is, if the rotational speed differences Df, Dr of the left and right wheels are smaller than the predetermined value Lm for both the front and rear wheels, the reliability flag Fa is set, while both judgments are made. Is YES, that is, one of the rotational speed differences Df and Dr between the front and rear wheels is the predetermined value Lm.
When it is larger, the reliability flag Fa is reset.

In FIG. 10, all the wheel speeds Wfl, Wfr, Wrl, Wrr of the four wheels are added in step S61, and it is determined in step S62 whether the added value W is zero. This determination ultimately determines whether or not the vehicle is stopped. When the vehicle is not stopped and is moving, the wheel speed W of the four wheels is further increased in step S63.
All of fl, Wfr, Wrl, and Wrr are integrated, and it is determined in step S64 whether the integrated value WM is zero. This determination ultimately determines whether or not at least one of the four wheels is in the locked state where the wheel speed is zero. When the vehicle is moving and none of the four wheels is in the locked state, the evaluation flag Fa is set in step S65 while the vehicle is stopped or at least one of the four wheels is in the locked state. Sometimes, the evaluation flag Fa is reset in step S66.

FIG. 11 shows a main flowchart of the rear wheel steering control in the vehicle four-wheel steering system according to the second embodiment of the present invention. Since the hardware configuration of the four-wheel steering system is the same as that of the first embodiment shown in FIGS. 1 and 2,
In the following description, reference numerals shown in these figures are used for description.

In FIG. 11, first, after waiting for the measurement timing in step S71, in step S72 the wheel speeds Wfr, Wfl, Wrr, W of the four wheels as vehicle motion information.
rl is measured by the wheel speed sensors 31 to 34, respectively, and the steering angle Fs as driver operation information is measured by the steering angle sensor 35 in step S73. continue,
In step S74, the rotational speed difference Rv between the left and right wheels, which is the difference between the wheel speeds Wrl and Wrr of the left and right driven wheels, is obtained at the addition point 41, and a filter calculation is performed by the filter 42 for the rotational speed difference Rv. . This rotation speed difference R
The filter calculation of v is performed by the formula described in the case of the first embodiment.

Then, in step S75, the normalization processing unit 4
At 3 the input information left / right wheel rotational speed difference Rv, vehicle speed V, steering angle Fs and steering speed Dfs are normalized, and in step S76 the neural network calculation unit 44 uses these normalized values as input variables. Neural network operation is performed to calculate the estimated yaw rate φ. The normalization and the neural network calculation are performed in the same manner as in the first embodiment.

Then, in step S78, the reliability judgment means 5
In step 2, the reliability of the estimated yaw rate φ is evaluated. For this reliability evaluation, for example, the rotational speed difference Dl, Dr between the front and rear wheels is calculated separately according to the flowchart shown in FIG. 8 as in the case of the first embodiment, and the rotational speed difference Dl between the front and rear wheels is calculated.
For example, the reliability of the estimated yaw rate φ is determined depending on whether Dr is larger or smaller than a predetermined value, and the determination is made based on the wheel speeds Wfl, Wfr, Wrl, and Wrr of the four wheels.

Then, in step S77, the reliability flag Fa
Is 1, that is, whether or not the estimated yaw rate φ is reliable, and if there is reliability, the rear wheels 3L, 3R are steered by H∞ control in step S79, while if not reliable, In step S80, steering of the rear wheels 3L and 3R is performed by map control. Then, the process returns to step S71.

The above H∞ control is performed as follows. That is, first, the target yaw rate φt of the vehicle is set based on the following equation, and the yaw rate deviation en between the target yaw rate φt and the estimated yaw rate φ is calculated.

[0053]

[Equation 1] However, A is the stability factor and L is the wheel base of the vehicle.

Thereafter, the steering control amount Rs of the rear wheels 3L, 3R as the feedback control amount of the H∞ control is determined based on the following vehicle state equation (1) and output equation (2) in the H∞ control.

[0055]

[Equation 2] However, Xh is a state quantity corresponding to a plurality of state quantities of the vehicle, and the state quantity of the vehicle is, for example, the sideslip angle of the vehicle, the steering angle of the rear wheels or its changing speed, the cornering force of the front wheels and the rear wheels, or the vehicle. Is the yaw rate that acts on, and these are estimated by both equations above.

Further, the control gains Acl, Bcl, Ccl and Dcl of the above-mentioned vehicle state equation (1) and output equation (2) are used.
Is calculated and determined in advance based on the control gain determination flow shown in FIG.

Next, the control gain determination flow in the H∞ control shown in FIG. 12 will be described.

In FIG. 12, first, in step S81, the performance target index W3 that gives stability when the characteristics of the vehicle change as the control gain characteristic of the frequency transfer function of the yaw rate change of the vehicle with respect to the steering of the rear wheels 3L, 3R is changed. The correction coefficient (weight) Wn for changing is changed according to the vehicle speed V. This change is made by determining the correction coefficient Wn to a large value in accordance with the increase in the vehicle speed V, as shown in step S81.

Then, in step S82, the performance target index W3 that gives the stability when the characteristics of the vehicle are changed is determined based on the following quadratic expression of the Laplace variable s.

[0060]

(Equation 3) However, a and c are constants. Based on the above equation, as shown in FIG. 13, the performance target index W3 that gives stability when the characteristics of the vehicle fluctuates is changed to a low frequency side when the correction coefficient Wn is large, and the correction coefficient Wn is small. When it is a value, it is changed to the high frequency side.

Then, in step S83, the rear wheel 3L,
The performance target index W1 (reciprocal number display) shown in FIG. 13 which gives the vehicle speed response and steady-state characteristics as the control gain characteristic of the frequency transfer function of the yaw rate change of the vehicle with respect to the 3R steering is expressed by the third order of the Laplace variable s below. Specify a fixed value based on an expression.

[0062]

[Equation 4] However, d, e, f, g and h are constants.

Then, in step S84, the above two performance target indexes W3 and W1 are converted into state equations, respectively, and both state equations and the following vehicle state equation (3) and output equation (4) are expressed. Each of the control gains Acl ~ Dcl of the H ∞ control is obtained by synthesizing the person into one state equation, decomposing the synthesized state equation into two Riccati equations, and solving this equation as an eigenvalue problem by the Hamilton matrix. Ask for.

[0064]

(Equation 5) In the vehicle state equation (3) and the output equation (4), β is the sideslip angle at the center of gravity of the vehicle, Cf and Cr are the cornering forces of the front and rear wheels, and Rs.
Is the steering angle of the rear wheels, lf and lr are the distances between the front and rear axles and the center of gravity of the vehicle, I is the moment of inertia of the vehicle with respect to yaw rate, m is the vehicle mass, V is the vehicle speed, Gy is the lateral elastic coefficient of the wheels, Kyf and Kyr is the cornering power of the front and rear wheels, respectively. Here, as the vehicle speed V, the actual vehicle speed detected by the vehicle speed detecting means 47 (see FIG. 2) is used, and the others are preset fixed values.

After that, in step S85, the control gains Acl to Dcl are changed from the continuous control system to the digital control system by the bilinear conversion method, and the process ends. The control gains Acl to Dcl can be determined by a method of sequentially reading the values stored in the map in advance according to the motion state of the vehicle, in addition to the above-described sequential calculation.

Further, in the map control of step S80 of FIG. 11, as shown in the same step, the steering control amount Rs of the rear wheels is calculated according to the following expression Rs = K × Fs according to the front wheel steering angle Fs. However, K is a proportional constant, and the proportional constant K is set to a large value in accordance with the vehicle speed up to a predetermined high vehicle speed value V0, and is gradually set to a small value at a vehicle speed exceeding the above predetermined value V0, and nonlinear It is designed to make the control system essentially stable in the area.

In short, in the case of the second embodiment, the control amount setting means 51 controls the estimated yaw rate φ of the vehicle, the estimated sideslip angle β of the vehicle, and the cornering forces Cf, Cr of the front and rear wheels.
The estimated yaw rate φ of the vehicle is feedback-controlled to the target yaw rate φt based on the vehicle state equation (1) and the output equation (2), where As the control gain characteristic of the frequency transfer function of the yaw rate change, two types of performance target index W3 that gives stability when the characteristic of the vehicle changes and performance target index W1 that gives the quick response and steady characteristics of the vehicle and the rear wheel steering angle Rs are Based on the input vehicle state equation (3) and the output equation (4), the vehicle motion characteristic (control gain characteristic) is a predetermined frequency (Fig. 13) at which the open loop transfer function becomes 0 db. 6r for 13
In the high frequency range exceeding (ad / sec), the performance target index W3 is provided immediately below the performance target index W3 that provides the above-mentioned stability, and in the low frequency range below a predetermined frequency (3 rad / sec), the performance target index W1 that provides the above steady-state characteristics is The control gains Acl to D of the above-mentioned state feedback control are positioned so as to be located immediately above.
It is designed to calculate cl. In addition, the regulation means 53
When the estimated yaw rate φ is unreliable, H
The control is stopped and the steering control amount Rs of the rear wheels 3L and 3R is set by map control.

Therefore, in the second embodiment described above,
In the case of normal control in which it is determined that the estimated yaw rate φ is reliable, the control amount setting means 51 calculates the estimated yaw rate φ of the vehicle, the estimated sideslip angle β of the vehicle, and the cornering forces Cf and Cr of the front and rear wheels. The estimated yaw rate φ of the vehicle is feedback-controlled to the target yaw rate φt based on the vehicle state equation (1) and the output equation (2), which are the vehicle state quantities, and at the same time, the yaw rate change of the vehicle with respect to the steering of the rear wheels 3L, 3R is performed. As a control gain characteristic of the frequency transfer function, a performance target index W3 that gives stability when the characteristic of the vehicle changes and a performance target index W1 that gives quick response and steady-state characteristics of the vehicle, and the rear wheel steering angle Rs are input. Based on the vehicle state equation (3) and the output equation (4), the stability is given in the high frequency range where the vehicle motion characteristics exceed a predetermined frequency. Control gain Acl ~ of the above-mentioned state feedback control so that it is located immediately below the performance target index W3 and above the performance target index W1 that gives the above-mentioned steady characteristics in the low frequency range below a predetermined frequency.
The H∞ control for calculating Dcl is performed. Therefore, FIG.
In contrast to the normal vehicle motion characteristics in which the H∞ control is not performed, the closed loop gain becomes a small value in the high frequency range exceeding the predetermined frequency (6 rad / sec), and becomes a large value in the low frequency range below the predetermined frequency. Become. As a result, in the high frequency range, the closed loop gain is small, so that the stability with respect to the characteristic variation of the vehicle is improved, the robust stability is improved, and in the low frequency range, a large value is obtained according to the steering wheel operation of the driver. The vehicle responds quickly with the closed loop gain, and the yaw rate with a very small deviation from the target yaw rate is generated in the vehicle, which improves the quick response and steady-state characteristics.

When it is determined that the estimated yaw rate φ is not reliable, the steering control amount Rs is changed by map control not related to the estimated yaw rate φ instead of the above H∞ control.
Is calculated, it is possible to prevent the control based on the incorrect estimated yaw rate φ, as in the case of the first embodiment.
It is possible to improve safety and the like.

In the first and second embodiments, in the yaw rate estimating means 44 for estimating the yaw rate, the neural network operation using the rotational speed difference Rv between the left and right rear wheels 3L and 3R which are driven wheels as an input variable. The estimated yaw rate φ was calculated by using the following method. Instead of this neural network calculation, a linear form derived from the geometrical relationship (so-called two-wheel model) during circular turning based on the rotational speed difference between the left and right driven wheels. It goes without saying that the yaw rate may be estimated.

[0071]

As described above, according to the vehicle control device of the present invention, the rotational speeds of all four wheels are detected, and the rotational speed difference between the left and right wheels is calculated or the yaw rate is estimated from these detected values. At the same time, the reliability of the estimation of the yaw rate is judged, and when it is not reliable, the control based on the estimated yaw rate is regulated. Therefore, the control based on the incorrect yaw rate estimation can be prevented, and the safety and the control can be ensured. It is possible to improve reliability. Particularly, it is effective when applied to the steering of the rear wheels in the four-wheel steering device as in the invention according to claim 2.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a four-wheel steering system for a vehicle according to a first embodiment of the present invention.

FIG. 2 is a block diagram of a control unit for rear wheel steering control.

FIG. 3 is a main flowchart of rear wheel steering control.

FIG. 4 is a diagram showing a map for setting a vehicle speed response coefficient.

FIG. 5 is a diagram showing a map for setting a steering angle response coefficient.

FIG. 6 is a flowchart of neural network calculation.

FIG. 7 is a flowchart of calculation of an intermediate layer transfer function.

FIG. 8 is a flowchart of reliability evaluation of an estimated yaw rate.

9 is a view corresponding to FIG. 8 showing a modified example.

FIG. 10 is a view corresponding to FIG. 8 showing another modification.

FIG. 11 is a view corresponding to FIG. 3 showing a second embodiment.

FIG. 12 is a flowchart for determining a control gain.

FIG. 13: Two performance target indicators W1 used for H∞ control
, W3.

FIG. 14 is a diagram showing a gain characteristic of a frequency transfer function of a motion characteristic of a vehicle for steering rear wheels when H∞ control is not performed.

[Explanation of symbols]

3L, 3R Rear Wheel 20 Rear Wheel Steering Device 30 Control Unit 31-34 Wheel Speed Sensor (Wheel Speed Detection Means) 41 Addition Point (Calculation Means) 44 Neural Network Calculation Unit (Yaw Rate Estimating Means) 52 Reliability Judgment Means 53 Regulation Means

Claims (5)

[Claims] 1. A vehicle control device for controlling a controlled object based on a yaw rate estimated from a rotational speed difference between the left and right wheels, the four wheel speed detecting means for detecting respective rotational speeds of the left and right front wheels and rear wheels. And a calculating means for obtaining the rotational speed difference between the left and right wheels from the rotational speed of the left and right front wheels or the rotational speed of the left and right rear wheels, and a yaw rate estimating means for estimating the yaw rate from the rotational speed difference between the left and right wheels obtained by the calculating means. A reliability determination means for determining the reliability of the yaw rate estimated by the estimation means, based on the rotation speed of the wheel used for the calculation of the left and right wheel rotation speed difference by the calculation means and the rotation speed of the other wheels, A control device for a vehicle, comprising: a restriction unit that restricts control based on the estimated yaw rate when the estimated yaw rate is not reliable as a result of the determination by the determination unit. 2. The control device for a vehicle according to claim 1, wherein the controlled object is a rear wheel steered when steering the front wheel. 3. The control amount of the rear wheel steering based on the yaw rate is determined by adding a vehicle speed sensitive control term having a vehicle speed as a factor and a yaw rate compensating term having a yaw rate as a factor. The vehicle control device according to claim 2, wherein the yaw rate compensation term is set to zero when the estimated yaw rate is not reliable. 4. The rear wheel steering control based on the yaw rate,
It is yaw rate feedback control that controls so that the actual yaw rate coincides with the target yaw rate at which the sideslip angle is zero.The regulation means stops the yaw rate feedback control when the estimated yaw rate is unreliable, and sets the vehicle speed in advance as a factor. The control device for a vehicle according to claim 2, wherein the steering of the rear wheels is controlled based on the map set as. 5. The reliability determining means obtains a rotational speed difference between the left front wheel and the left rear wheel and a rotational speed difference between the right front wheel and the right rear wheel, and when the rotational speed difference is smaller than a predetermined value, respectively. The vehicle control device according to claim 1, wherein the yaw rate is determined to be reliable, and the yaw rate is determined to be unreliable when at least one of the rotational speed differences is greater than a predetermined value.