JP4295138B2 - Steering control device - Google Patents

Description

  The present invention relates to a steering control device that obtains a steering assist torque that cancels a yaw rate change of a vehicle by using image information of a traveling path of the vehicle and performs steering control.

Conventionally, in order to improve the running stability (feeling of wobbling) of a vehicle, it is possible to detect the movement in the rotational direction of the vehicle, that is, the yaw rate (the rate of change of the yaw angle), and perform the steering control so as to cancel it. It is described in Patent Document 1 and Patent Document 2.
JP-A-6-92254 Japanese Patent Laid-Open No. 11-147483

  In the vehicle steering device disclosed in Patent Document 1, the vehicle behavior is recognized based on the yaw rate detection value from the yaw rate sensor, and the reaction force component of this behavior is applied to the steering device, so that the vehicle is driven by a crosswind or the like. Suppresses unstable behavior.

  However, the yaw rate detection value detected by the yaw rate sensor includes yaw due to a driver's steering operation in addition to an element that causes an unstable behavior in the vehicle such as a crosswind when the vehicle is running on a curve.

  For this reason, the reaction force component includes a component that cancels the steering operation by the driver in addition to a component that suppresses the unstable behavior of the vehicle due to a crosswind or the like. .

  Patent Document 2 describes a method of detecting the signal of the yaw rate sensor and driving the electric power steering device using this signal as a steering assist torque. In this method, the curve curvature and the vehicle speed are obtained when passing the curve road. Since the reference yaw rate, that is, the DC component, is detected, if this is fed back to the steering torque, an undesirable (heavy curve) steering feeling will result.

DC component of the yaw rate is generally large enough to the relative yaw component (wander component) In addition, since the relative yaw component present in a relatively low frequency band below IH _Z, also the S / N ratio by using the band-pass filter Therefore, it is difficult to extract the relative yaw rate.

  On the other hand, a method has been proposed in which the amount of change in the forward gazing point displacement is calculated as a relative yaw rate component (staggered component) from an image of a CCD camera or the like fixed to the vehicle. Since this method is a method for detecting the rotational movement with respect to the lane of the vehicle fixed coordinate system, it can accurately extract the relative yaw rate component excluding the DC component generated when passing through a curve, etc. The true value of the yaw rate (flicker component) cannot be extracted.

  That is, as shown in FIG. 6, when the vehicle 50 goes straight with a certain yaw angle, DC yaw rate = 0, relative yaw rate = 0, CCD camera estimated yaw rate ≠ 0, and the true value of the relative yaw rate is extracted. Can not.

  Also, as shown in FIG. 7, when the vehicle 50 travels on a curved road without following the curve curvature, DC yaw rate ≠ 0, relative yaw rate = 0, CCD camera estimated yaw rate ≠ 0, and the true value of the relative yaw rate cannot be extracted. .

  For this reason, in the past, the DC component was cut using a high-pass filter for the estimated relative yaw rate estimated from the camera image, but the high-pass filter inevitably promotes the noise component, and is used for control purposes. It is not preferable for use as a signal.

  The present invention has been made in view of the above points, and an object thereof is to extract only a wobbling component caused by vehicle motion, that is, a relative yaw rate, without performing a filtering process that promotes a noise component, so that the steering feeling is uncomfortable. An object of the present invention is to provide a steering control device capable of performing steering assist without any trouble.

  In order to solve the above-described problem, the invention according to claim 1 is a steering control device having an actuator for turning a steering wheel of a vehicle, for example, a traveling path in the vehicle traveling direction is detected as image data by a CCD camera, From the detected image data, at least three forward gazing points with different distances from the vehicle on the vehicle traveling direction line are taken as sample data, and a reference line along the travel path is based on the taken sample data A current yaw angle of the vehicle is calculated, and a width direction displacement estimation amount representing a deviation in the width direction of the vehicle with respect to the reference line at a predetermined forward gazing point is obtained based on the acquired sample data, and the width The gazing point change component caused by the current yaw angle is removed from the estimated directional displacement amount to calculate the vehicle yaw rate, and the calculated yaw rate is calculated. Calculating a steering assist torque to extinguish Chi, on the basis of the calculated steering assist torque, and configuring the actuator to control the drive.

  According to the first aspect of the present invention, since the yaw rate calculation means removes the gazing point change component caused by the current yaw angle from the estimated displacement in the width direction, the driver's reflection reflected in the yaw angle. The yaw rate due to the steering operation is canceled, and only the relative yaw rate component due to disturbances such as cross winds and road surface irregularities is accurately extracted.

  For this reason, the steering assist torque calculating means can obtain an assist torque that cancels only the relative yaw rate due to the disturbance, thereby suppressing the behavior of the vehicle due to the disturbance, and feeling the uncomfortable feeling of steering that cancels the steering operation. It does not occur.

  Further, the relative yaw rate caused by the vehicle motion can be extracted without passing through a secondary signal processing means such as a bandpass filter or a high-pass filter to promote noise components.

  According to a second aspect of the present invention, the yaw rate calculation means is configured to simultaneously obtain a yaw rate of the vehicle at a plurality of forward gazing points and average the plurality of yaw rates.

  According to the second aspect of the invention, since a plurality of relative yaw rate calculations are performed simultaneously, noise components generated from the resolution of the camera are suppressed, and a more stable relative yaw rate signal can be obtained without a time delay. Can do.

  Embodiments of the present invention will be described below with reference to the drawings. Although the two phenomena described in FIGS. 6 and 7 seem to be completely different, they have a common point in that “the yaw angle with respect to the lane is not zero”. In other words, “go along the curve” = “run with the yaw angle kept at 0”.

  Therefore, in the present invention, the amount of displacement of the forward gazing point and the yaw angle are simultaneously estimated from the camera data, and the relative yaw rate is estimated by compensating for the displacement component created by the yaw angle among the displacement components of the forward gazing point. Only the wobbling component caused by the vehicle motion is extracted without passing through the processing.

  FIG. 1 shows an overall configuration of a system in which the present invention is applied to an electric power steering apparatus, and FIG. 2 shows a control flow of the apparatus of FIG.

  In FIG. 1, a steering wheel 1 is connected to a steering shaft 2, and the steering shaft 2 is connected to a connecting shaft 4 via a universal joint 3. The connecting shaft 4 is connected to a pinion 5 a of a rack and pinion type steering gear 5, and the pinion 5 a meshes with the rack 6.

  Therefore, the rotational motion input from the steering wheel 1 is converted into the reciprocating motion of the rack 6 via the pinion 5a, and the two steering wheels (front wheels) 8 are steered in a desired direction via the tie rod 7.

  An electric motor 9 is disposed on the pinion 5a and serves as an actuator for turning the steered wheels 8. A torque sensor 10 outputs a signal corresponding to the direction and magnitude of the steering force (steering torque) input by the driver. Reference numeral 11 denotes a camera as travel path detection means for imaging a travel path in the vehicle traveling direction.

  Reference numeral 12 denotes an arithmetic unit for calculating the steering assist torque, comprising the sample data fetching means, the current yaw angle calculating means, the yaw rate calculating means, the steering assist torque calculating means, and the means for estimating the vehicle state and the road state. is there.

  The arithmetic unit 12 receives a steering torque detection signal of the torque sensor 10, image data of the camera 11, a vehicle speed detection signal from a vehicle speed detection unit (not shown), and the like.

  Reference numeral 13 denotes a motor drive circuit as drive control means for driving and controlling the electric motor 9 based on the steering assist torque calculated by the calculation unit 12.

  In FIG. 2, reference numeral 21 denotes a travel path detection unit configured by the camera 11 of FIG. 1, for example. Reference numeral 22 denotes sample data fetching means for fetching at least three forward gazing points at different distances from the vehicle on the vehicle traveling direction line as sample data from the image data detected by the travel path detection means 21.

  Reference numeral 23 denotes a yaw for estimating and calculating a vehicle deflection angle representing the inclination of the vehicle with respect to a reference line (for example, the center of the lane) along the traveling path, that is, a yaw angle, based on the sample data captured by the sample data capturing means 22 Angle estimation means (current yaw angle calculation means).

  24 is based on the vehicle speed detected by the vehicle speed detection means 25 such as a vehicle speed sensor, the yaw angle calculated by the yaw angle estimation means 23, and the sample data captured by the sample data capture means 22. Relative yaw rate estimating means (yaw rate calculating means) for calculating a yaw rate (yaw angle change rate).

  26 calculates a steering assist torque that cancels the relative yaw rate based on the relative yaw rate calculated by the relative yaw rate estimating means 24 and the steering torque detected by the steering torque detecting means (for example, the torque sensor 10 in FIG. 1) 27. Steering assist torque calculating means.

  Reference numeral 28 denotes drive control means for drivingly controlling the electric motor 9 as an actuator based on the steering assist torque calculated by the steering assist torque calculating means 26.

  Next, the operation of the apparatus configured as described above will be described with reference to FIG. FIG. 3 shows the estimated amount and the calculation amount when the vehicle travels on a straight road. The vehicle is inclined at an angle of ψ with respect to the X coordinate of the vehicle fixed coordinate system X, Y.

  First, the definition of each data used for the estimation and calculation will be described.

  Ref is a reference line along the road, for example, the center line of the lane.

  Vx is the vehicle speed.

a ₁ is a forward gazing point as seen from the current vehicle position.

a ₂ is a forward gazing point as seen from the vehicle position after a predetermined time Δt has elapsed.

  L is the distance [m] on the X-axis from the position where the vehicle is present to the forward gazing point.

w ₀ is a width direction displacement estimated value [m] with respect to the reference line Ref at the forward gazing point a ₁ .

wΔt is a width direction displacement estimated value [m] with respect to the reference line Ref at the forward gazing point a ₂ .

  ψ is the current yaw angle [rad] of the vehicle with respect to the reference line Ref.

  ψ′Δt is the yaw angle [rad] corresponding to the yaw rate caused by the disturbance.

y ₀ is a width direction displacement amount [m] representing a width direction shift with respect to the reference line Ref at the current position of the vehicle.

  yΔt is a width direction displacement amount [m] representing a width direction deviation with respect to the reference line Ref at the vehicle position after the predetermined time Δt has elapsed.

First, the width direction displacement estimated value w _{0 at} the forward gazing point a ₁ viewed from the current position (t = 0) of the vehicle is expressed by the following equation (1).

w ₀ = y ₀ + L tan ψ (1)
In general, since the yaw angle ψ is very small, w ₀ is expressed by the following equation (2).

w ₀ = y ₀ + Lψ (2)
Next, the displacement yΔt in the width direction of the vehicle at t = Δt is established by the three terms of the following equation (3).

yΔt = y ₀ + (VxΔt) tan ψ + (VxΔt) tan (ψ′Δt) (3)
In Expression (3), y ₀ represents an offset displacement component, (VxΔt) tan ψ represents a displacement component due to the current yaw angle ψ, and (VxΔt) tan (ψ′Δt) represents a displacement component due to the yaw rate (vehicle motion). Yes.

If this equation (3) is also approximated as above,
yΔt = y ₀ + (VxΔt) + (ψ + ψ′Δt) (4)
It becomes.

The estimated width direction displacement value wΔt at the gazing point a _{2 as} viewed from the vehicle position at t = Δt is:
wΔt = yΔt + Ltan (ψ + ψ′Δt) = y ₀ + (VxΔt + L) (ψ + ψ′Δt) (5)
Is required.

  When the equation (2) is subtracted from the equation (5), the generated yaw rate ψ ′ is expressed as the following equation (6).

  As can be seen from this equation (6), the yaw rate ψ ′ is VxΔtψ, that is, the displacement component due to the yaw angle ψ described in equation (3) (the gazing point changing component due to the current yaw angle) is removed, and the yaw rate due to disturbance The minute (ψ′Δt component) is accurately calculated.

  For this reason, the yaw rate due to the steering operation of the driver reflected in the yaw angle ψ is canceled, and only the relative yaw rate component due to disturbance such as cross wind and road surface unevenness is accurately obtained.

  The relative yaw rate (ψ ′) thus determined is multiplied by the control gain Kp in the steering assist torque calculating means 26, and further, dead zone processing, limiter processing, etc. are performed to determine the steering assist torque Tk. The drive control means 28 creates a current command signal according to the determined steering assist torque Tk, and the electric motor 9 is controlled by the current command signal.

  The calculation unit 12 shown in FIG. 1 performs various estimation amounts and calculation amounts such as ψ required for the yaw rate calculation of the present invention. How to obtain such information will be described below.

  FIG. 4 shows the estimated amounts and calculation amounts on a straight road, and FIG. 5 shows the estimated amounts and calculation amounts on a curved road.

  4 and 5, a relative coordinate system in which the current vehicle position in the fixed coordinate systems X and Y is X = 0, the center of the lane is Y = 0, the vehicle traveling direction is the x axis, and the direction orthogonal thereto is the y axis. It is said.

  First, the definition of each data used for the estimation and calculation will be described.

  Ref is a reference line along the traveling road, and is the center of the lane in this embodiment.

  Vx is the vehicle speed.

  a, b, and c are at least three forward gazing points on the vehicle traveling direction line x, each having a different distance from the vehicle. At least three forward gazing points used as the sample data are different from the first pair of forward gazing points (for example, a and b) indicating two points having different forward distances from the vehicle, and the forward distances from the vehicle are different 2 A point is shown, and at least one of the two points is constituted by a second pair of front gaze points (for example, b, c) different from the first pair of front gaze points.

  La is the distance [m] on the X-axis from the vehicle to the forward gazing point a.

  Lb is the distance [m] on the X-axis from the vehicle to the forward gazing point b.

  Lc is the distance [m] on the X axis from the vehicle to the forward gazing point c.

  wa is a displacement amount [m] in the width direction with respect to the reference line Ref at the front gazing point a.

  wb is a displacement amount [m] in the width direction with respect to the reference line Ref at the front gazing point b.

  wc is a displacement amount [m] in the width direction with respect to the reference line Ref at the forward gazing point c.

ψ _ab is an estimated value of the vehicle deflection angle with respect to the reference line Ref, that is, an estimated value for lane yaw angle [rad], and is a ratio (wb−) of the deviation between the width direction displacements wa and wb and the deviation between the distances La and Lb. wa) / (Lb−La).

ψ _bc is an estimated value of the vehicle deflection angle with respect to the reference line Ref, that is, an estimated value for lane yaw angle [rad], and is a ratio (wc−) of the deviation between the width direction displacement amounts wb and wc and the deviation between the distances Lb and Lc. wb) / (Lc−Lb).

w _0ab is a width direction displacement estimated value [m] at the present time of the vehicle, which is obtained from the width direction displacement amounts wa and wb.

w _0bc is an estimated width direction displacement value [m] of the vehicle at the present time obtained from the width direction displacement amounts wb and wc.

y ₀ is a width direction displacement amount [m] representing a width direction shift with respect to the reference line Ref at the current position of the vehicle.

  ρ is the curvature [1 / m] of the road with the left curve being positive.

  ψ is the vehicle yaw angle (vehicle deflection angle) and represents the inclination of the fixed coordinate system with respect to the X axis.

w _Tab is a width direction displacement estimated value [m] at the front gazing point for control after T seconds, which is obtained from the wa and wb.

w _Tbc is the width direction displacement estimated value [m] at the front gazing point for control after T seconds, which is obtained from wb and wc.

w _T is a control deviation at the forward gazing point for control after T seconds, and indicates a width direction displacement amount representing a deviation in the width direction of the vehicle with respect to the reference line Ref.

(1) Estimation of road condition / vehicle condition on a straight road In FIG. 4, when traveling on a straight road, a displacement wa in the width direction with respect to the center of the lane at the gazing points a, b, c ahead of the vehicle, that is, the reference line Ref, by camera image output , Wb, wc are obtained, the state of the vehicle with respect to the current lane is expressed as follows using two of these points.

The lane yaw angle estimated value ψ _ab is expressed by the following equation (7).

Since this yaw angle is generally small, ψ _ab is expressed by the following equation (8).

Next, the width direction displacement estimated value w _0ab at the present time and the width direction displacement estimated value w _Tab after T seconds are obtained geometrically as follows.

w _0ab = wa−ψ _ab · La (9)
w _Tab = wa−ψ _ab (VxT−La) (10)
T is a time constant determined in consideration of the delay time for calculating displacement data from the image and the response of the vehicle.

By using w _Tab calculated as described above as a control deviation, lane / keep control on a straight road becomes possible.

  The calculation formula when using the data of the forward gazing point b, c or c, a is the following formulas (8 ′), (8〃), (9 ′), (9〃). Since the yaw angle with respect to lane is the same, the equations (8 ′) and (8〃) have the same value as the equation (8), and the equations (9 ′) and (9〃) are the same as the equation (9). Value.

w _0bc = wc−ψ _bc · Lc (9 ′)
w _0ac = wc−ψ _ac · Lc (9〃)
(2) Estimation of road condition / vehicle condition on curved road In order to perform lane keeping control on a curved road as shown in FIG. 5, active assistance according to the curve curvature is required. For this reason, it is necessary to give the road state / vehicle state as control information. In this embodiment, the width direction displacements wa, wb, wc (from the center of the lane received from the camera are used without using the conventional steering angle sensor. (3 or more points) data is used to estimate the current road state curvature ρ, width direction displacement y ₀ , and vehicle yaw angle ψ.

First, consider a fixed coordinate system XY in which the current vehicle position is X = 0, the lane center is Y = 0, and the X axis is a tangent to the lane center (reference line Ref) at the origin. The curvature ρ is constant and sufficiently small, and the displacement amount Y at the center of the lane due to the curvature in the coordinate system XY is:
^{Y = ρX 2/2 ... (} 11)
Is approximated by Thus, the width-direction displacement amount w _{T with} respect to the center of the lane (reference line Ref) at the forward gazing point (control target position) after T seconds (viewed from the vehicle) according to the vehicle coordinate system is expressed by the following equation.

If it is a straight road, the value of ψ _ab , w _0ab , w _Tab does not change even if any combination of discrete displacement data (wa, wb, wc) is used, and w _0ab = y ₀ , ψ _ab = ψ is there. On the other hand, on the curved road, as shown in FIG. 4, the calculation result (estimated value) differs depending on the combination of sampling points, and none of them agrees with the true value.

Here, the deviation of the vehicle position prediction point calculated from two different points will be considered. This vehicle position deviation (current width direction estimated value, for example, w _0ab ) is _found from the difference in inclination, that is, the curvature ρ, from Equation (9). Further, since it is assumed that the curvature ρ is sufficiently small, it can be said that the vehicle position deviation and the curvature ρ have a linear function relationship.

  This primary relationship is obtained by the following procedure. Assuming that the vehicle is at the origin of the XY coordinates with the vehicle yaw angle ψ = 0, the displacement wa in the width direction seen from the vehicle by the curvature ρ at an arbitrary forward gazing point, for example, a (distance La from the origin) is ) Is obtained by the following equation.

^{wa = -ρLa 2/2 ... (} 13)
Similarly, the width direction displacement amount wb at the front gazing point b is
^{wb = -ρLb 2/2 ... (} 13 ')
Similarly, the displacement amount wc in the width direction at the forward gazing point c is
wc = -ρLc ^2/2 ... (13〃)
It becomes.

Here, the width direction displacement estimated value w _0ab at the present time by arbitrary La and Lb is obtained by substituting Equation (13) into Equation (9).

Similarly, the width direction displacement estimated value w _{0bc at the} present time by arbitrary Lb and Lc is

It becomes. Therefore, the curvature ρ is expressed as a linear function of the vehicle position estimated value deviation (w _0bc −w _0ab ) as shown in the following equation from the equations (15)-(14).

  As an example, if La = 10 [m], Lb = 20 [m], and Lc = 30 [m], the curvature ρ is expressed by the following equation.

ρ = (w _0bc −w _0ab ) / 200 (17)
Next, an estimated lane yaw angle value ψ _ab obtained from two arbitrary points will be considered. Since this is considered that the influence of the vehicle yaw angle ψ is added to the inclination ψ _ρ created by the curvature ρ, the following equation is established.

ψ _ab = ψ _ρ + ψ (18)
Here, ψ _ρ is obtained from Equation (8) and Equation (13).
ψ _ρ = −ρ (La + Lb) / 2 (19)
It is expressed. Thus, the vehicle yaw angle ψ is obtained as a linear function of the curvature ρ as shown in the following equation.

As an example, if La = 10 [m] and Lb = 20 [m],
ψ = ψ _ab + 15ρ (21)
Is required.

Finally, the vehicle position estimated value at any two points, that is, the vehicle width direction displacement estimated value w _0ab at the present time is set to the width direction displacement amount y ₀ from the center of the vehicle lane (reference line Ref) and the curvature ρ. It can be said that the displacement error component w _{0ρ produced by} is added, and is represented by the following.

w _0ab = y ₀ + w _0ρ (22)
By substituting equation (13) into equation (9), w _0ρ is obtained, and equation (22) is summarized as a linear function of curvature ρ as follows.

As an example, if La = 10 [m] and Lb = 20 [m],
y ₀ = w _0ab −100ρ (24)
It becomes.

Based on the width direction displacements wa, wb, wc (three or more points) from the center of the lane (reference line Ref), the current road condition, that is, the curvature ρ, the width direction displacement y ₀ , A vehicle yaw angle ψ is determined.

  Considering the forward gazing point as a fixed value, various quantities to be obtained are expressed by addition / subtraction of the width direction displacement (wa, wb, wc) data as follows. Here, an example in the case of La = 10 [m], Lb = 20 [m], and Lc = 30 [m] is shown.

That is, when ψ _ab = (wb−wa) / (Lb−La) in equation (8) is substituted for w _0ab = wa−ψ _ab · La in equation (9),
w _0ab = 2wa−wb (25)
Similarly, when ψ _bc = (wc−wb) / (Lc−Lb) in the equation (8 ′) is substituted for w _0bc = wc−ψ _bc · Lc,
w _0bc = 3wb-2wc (26)
Similarly, if ψ _ac = (wc−wa) / (Lc−La) in the equation (8 （) is substituted for w _0ac = wc−ψ _ac · Lc,
w _0ac = (3wa / 2) − (wc / 2) (27)
It becomes.

  And combining these equations (25), (26), (27),

  It becomes.

Therefore, substituting w _0ab and w _0bc in the equations (25) and (26) into the equation (16),
ρ = 10 ⁻² × (−wa + 2wb−wc) (31)
Is required.

Also, substituting ψ _{ab in} equation (8) and ρ in equation (31) into equation (20),

  Is required.

Also, by substituting w _{0ab in the} equation (25) and ρ in the equation (31) into the equation (23),
y ₀ = 3wa-3wb + wc (33)
Is required.

  Since wa, wb, and wc used in the above equations (31), (32), and (33) are parameters obtained from the sample data of the camera image, the steering assist torque Tk can be obtained by extremely simple calculation.

  The sample data may be one of gazing point data used for yaw angle calculation, or may be sample data separate from the yaw angle calculation. The displacement change amount (width direction displacement amount) can be calculated by differential calculation or differential calculation.

  Alternatively, a plurality of samples for obtaining the change amount may be provided, and the relative yaw rate may be averaged after calculating the relative yaw rate at the same time. When configured in this manner, it is possible to suppress a noise component generated from the resolution of the camera and obtain a more stable relative yaw rate signal without a time delay.

  In the present embodiment, since the current yaw angle ψ is used for the yaw rate calculation, accurate yaw rate calculation can be performed even when the traveling road and the vehicle are not parallel.

  Further, since the steering assist torque calculating means 25 takes in the steering torque component detected by the steering torque detecting means 26 as data, the steering assist torque calculating means 25 can calculate the reaction torque considering the driver's steering torque. Steering assist can be performed with no sense of incongruity in the ring.

The block diagram of the embodiment which applied this invention to the electric power steering apparatus. The block diagram which shows one embodiment of this invention. Explanatory drawing showing the state of the estimated value of this invention and the amount of calculations in a straight road. Explanatory drawing showing the state of the estimated value of this invention and the amount of calculations in a straight road. Explanatory drawing showing the estimated value of this invention and the amount of calculations in a curved road. Explanatory drawing which shows the problem at the time of the straight road driving | running | working in the conventional steering device. Explanatory drawing which shows the problem at the time of curve road driving | running | working in the conventional steering device.

Explanation of symbols

8 ... steering wheel 10 ... torque sensor (steering torque detecting means)
11 ... Camera (traveling path detection means)
12 ... arithmetic unit 13 ... motor drive circuit (drive control means)
21: Traveling path detection means 22 ... Sample data capturing means 23 ... Yaw angle estimation means (current yaw angle calculation means)
24: Relative yaw rate estimation means (yaw rate calculation means)
25 ... Vehicle speed detecting means 26 ... Steering assist torque calculating means 27 ... Steering torque detecting means 28 ... Drive control means

Claims (2)

In a steering control device having an actuator for turning a steering wheel of a vehicle,
A road detection unit for detecting a road in the vehicle traveling direction as image data;
Sample data capturing means for capturing, as sample data, at least three forward gazing points at different distances from the vehicle on the vehicle traveling direction line from the image data detected by the travel path detecting means;
A current yaw angle calculating means for calculating a current yaw angle of the vehicle with respect to a reference line along the traveling path, based on the sample data acquired by the sample data acquiring means;
Based on the sample data captured by the sample data capturing means, a width direction displacement estimation amount representing a shift in the vehicle width direction with respect to the reference line at a predetermined forward gazing point is obtained, and the current width direction displacement estimation amount is obtained from the width direction displacement estimation amount. A yaw rate calculating means for calculating a yaw rate of the vehicle by removing a gazing point changing component caused by the yaw angle;
Steering assist torque calculating means for calculating a steering assist torque that cancels the yaw rate calculated by the yaw rate calculating means;
A steering control device comprising: drive control means for drivingly controlling the actuator based on the steering assist torque calculated by the steering assist torque calculating means. The steering control device according to claim 1, wherein the yaw rate calculation means simultaneously obtains the yaw rate of the vehicle at a plurality of forward gazing points and averages the plurality of yaw rates.

Images