JP5332703B2 - Lane maintenance support device and lane maintenance support method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for supporting lane keeping effectively preventing deviation from a traveling lane, while reducing sense of incongruity caused by deviation from a traveling line intended by a driver. <P>SOLUTION: Lateral displacement reference positions LXL, LXR are provided to each of right and left in the width direction from a center in the width direction in the traveling lane L where one's own vehicle travels. When the own vehicle is positioned at least between the right and left lateral displacement reference positions LXL, LXR, feedback control is performed on the own vehicle to reduce yaw angle deviation. When the own vehicle is outside the right or left lateral displacement position LXL, LXR with respect to the center in the traveling lane, the feedback control is performed to reduce the above angel deviation and lateral displacement deviation. A phase of a rear wheel turning direction is controlled relatively with respect to a front wheel turning direction in accordance with at least one of the angle deviation and the lateral direction deviation. <P>COPYRIGHT: (C)2010,JPO&amp;INPIT

Description

  The present invention relates to a lane keeping assist device and a lane keeping assist method for preventing a host vehicle from traveling along a traveling lane and deviating from the traveling lane.

As a lane keeping assist device, there is an invention described in Patent Document 1, for example.
The technique described in Patent Document 1 is a technique for controlling the turning angle of a wheel so that the angle deviation between the traveling direction of the host vehicle and the traveling lane becomes small. This is intended to prevent the host vehicle from deviating from the traveling lane.

Japanese Patent No. 3729494

  As described in Patent Document 1, when the traveling direction of the host vehicle is simply controlled so that the angle deviation with respect to the traveling lane becomes zero, the angle deviation is on the departure side and the angle deviation on the departure avoidance side. The same control will intervene when there is a mark. For this reason, there is a trade-off relationship between the departure prevention effect when the angle deviation occurs toward the departure side and the control intervention discomfort when the angle deviation occurs toward the departure avoidance side.

Moreover, the said prior art only controls the advancing direction of the own vehicle so that an angle deviation may be set to zero. For this reason, when the deviation from the driving lane of the own vehicle cannot be prevented by the above control, there is the following problem. That is, although the host vehicle deviates from the driving lane, the control ends when the host vehicle becomes parallel to the driving lane, and the host vehicle is returned to the driving lane, or within the driving lane. There is a problem that the effect of fastening the host vehicle is insufficient.
The present invention pays attention to the above points, and provides a lane keeping assist device capable of effectively preventing deviation from a driving lane while reducing a sense of incongruity resulting from a deviation from a driving line intended by a driver. The issue is to provide.

In order to solve the above-described problems, the present invention provides a lateral displacement threshold value in a travel lane in which the host vehicle travels. When the host vehicle passes the lateral displacement threshold value from the center side of the traveling lane, the traveling direction of the host vehicle is controlled so that the angle deviation with respect to the traveling lane is smaller on the traveling lane center side than the lateral displacement threshold value. Thus, on the outer side in the width direction of the travel lane with respect to the lateral displacement threshold, the traveling direction of the host vehicle is controlled so as to reduce at least the lateral displacement deviation from the lateral displacement threshold.
At this time, the phase of the steering direction of the rear wheels relative to the steering direction of the front wheels is controlled in accordance with at least one of the angular deviation and the lateral deviation.

According to the present invention, when the host vehicle is located on the center side of the travel lane, the host vehicle follows the travel lane according to the travel line intended by the driver by controlling the angle deviation to be small. And run. On the other hand, when the host vehicle is located on the end side of the traveling lane, the control for returning to the central portion of the traveling lane intervenes so that the vehicle can be appropriately kept in the traveling lane.
Furthermore, control convergence can be improved by controlling the phase of the steering direction of the rear wheels relative to the front wheels in accordance with the angular deviation and the lateral deviation.
As described above, according to the present invention, it is possible to effectively prevent the departure from the driving lane while reducing the uncomfortable feeling resulting from the deviation from the driving line intended by the driver.

It is a figure explaining the system configuration of the vehicle concerning the embodiment based on the present invention. It is a figure explaining the process of the lane maintenance assistance controller which concerns on 1st Embodiment based on this invention. It is a top view explaining the relationship of each value which concerns on 1st Embodiment based on this invention. It is a top view explaining the relationship of each value which concerns on 1st Embodiment based on this invention. It is a figure which shows the relationship between the lateral displacement X and a lateral displacement deviation. It is a figure which shows the relationship between yaw angle (theta) and yaw angle deviation. It is a conceptual diagram which shows a curve IN side gain map. It is a conceptual diagram which shows a curve OUT side gain map. It is a conceptual diagram which shows the state of feedback gain Ky_R and Ky_L. It is a figure which shows the characteristic of the map of the correction gain Key for rear-wheel steering gain calculation. It is a figure which shows the characteristic of the map of the correction gain Kel for rear-wheel steering gain calculation. It is a figure which shows the characteristic of the map of correction | amendment gain Kvy for rear-wheel steering gain calculation. It is a figure which shows the characteristic of the map of correction | amendment gain Kvl for rear-wheel steering gain calculation. It is a figure which shows the characteristic of the map of correction | amendment gain Kdy for rear-wheel steering gain calculation. It is a figure which shows the characteristic of the map of correction | amendment gain Kdl for rear-wheel steering gain calculation. It is a figure which shows the value of a horizontal position and a weighting coefficient. It is a figure explaining the operation | movement of lateral position deviation predominance which concerns on 1st Embodiment based on this invention. It is a figure explaining operation of yaw angle deviation predominance concerning a 1st embodiment based on the present invention. It is a figure explaining the operation | movement which concerns on 1st Embodiment based on this invention. It is a figure explaining the operation | movement in the curve road which concerns on 1st Embodiment based on this invention. It is a conceptual diagram which shows the driving | running locus | trajectory at the time of deviation which concerns on 1st Embodiment based on this invention. It is a top view explaining the relationship of each value which concerns on the modification of 1st Embodiment based on this invention. It is a figure which shows the weighting of the departure side transition area | region which concerns on 1st Embodiment based on this invention.

Next, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a system schematic configuration diagram of a host vehicle to which the lane keeping assist device of the present embodiment is applied.
The host vehicle of this embodiment employs a steer-by-wire system.
(Constitution)
First, the configuration will be described with reference to FIG.
A steering input shaft 30 is connected to the steering wheel 12 operated by the driver. The steering input shaft 30 is provided with a handle angle sensor 1 that detects the steering angle of the steering wheel 12. The handle angle sensor 1 outputs the detected steering angle signal to the steering controller 11.
A first intermediate shaft 31 is connected to the steering input shaft 30 via the steering torque sensor 2. The steering torque sensor 2 detects the steering torque input to the steering input shaft 30 and outputs the torque signal to the steering controller 11.

  The steering reaction force actuator 3 is connected to the first intermediate shaft 31. The steering reaction force actuator 3 applies a steering reaction force to the first intermediate shaft 31 based on a command from the steering controller 11. A steering reaction force motor angle sensor 4 is provided on the steering reaction force motor of the steering reaction force actuator 3. The steering reaction force motor angle sensor 4 detects the rotational angle position of the steering reaction force motor and outputs the detection signal to the steering controller 11.

  A second intermediate shaft 32 is connected to the first intermediate shaft 31 via the mechanical backup device 10. The mechanical backup device 10 is in a state in which torque transmission between the first intermediate shaft 31 and the second intermediate shaft 32 is cut off in a normal state. Further, the mechanical backup device 10 connects the first intermediate shaft 31 and the second intermediate shaft 32 based on a command from the steering controller 11 to enable torque transmission.

  The second intermediate shaft 32 is connected to the steering output shaft 33 via the steering torque sensor 7. The steering actuator 5 is connected to the second intermediate shaft 32. The turning actuator 5 rotates and displaces the second intermediate shaft 32 based on a command from the steering controller 11. A steering actuator angle sensor 6 is provided in the steering motor of the steering actuator 5. The steered actuator angle sensor 6 detects the rotational angle position of the motor of the steered actuator 5 and outputs a detection signal to the steering controller 11.

  The steering output shaft 33 is connected to the rack shaft 34 via a rack and pinion mechanism. That is, the pinion connected to the steering output shaft 33 meshes with the rack of the rack shaft 34. The rack shaft 34 is disposed with its axis directed in the vehicle width direction. Then, by rotating the steering output shaft 33, the rack shaft 34 is displaced in the axial direction toward the vehicle width direction. Reference numeral 8 denotes a pinion angle sensor 8 that detects the rotation angle of the pinion and outputs it to the steering controller 11.

  The left and right ends of the rack shaft 34 are connected to the knuckle via the left and right tie rods 35, respectively. Reference numeral 36 denotes a knuckle arm protruding from the knuckle. The knuckle rotatably supports the front wheel 13 which is a steering wheel. The tie rod axial force sensor 9 is provided on the tie rod 35. The tie rod axial force sensor 9 detects the axial force of the tie rod 35 and outputs a detection signal to the steering controller 11.

Reference numeral 25 denotes a rear wheel steering actuator. The rear wheel steering actuator 25 axially displaces the rack shaft 50 in the vehicle width direction in response to a signal from the steering controller 11. The left and right end portions of the rack shaft 50 are connected to the knuckle for the rear wheel via the left and right tie rods 51, respectively. Reference numeral 52 denotes a knuckle arm protruding from the knuckle for the rear wheel. The knuckle supports the rear wheel 40 rotatably.
In addition, the host vehicle state parameter 14 is input to the steering controller 11. The own vehicle state parameter 14 is, for example, a vehicle speed detected by the vehicle speed detecting means or a friction coefficient estimated value of the running road surface detected by the road surface friction coefficient estimating means.

  The steering controller 11 outputs a steering command value with a steering angle corresponding to the steering angle detected by the steering wheel angle sensor 1 to the steering actuator 5 and outputs a command value for applying a steering reaction force. Output to the force actuator 3. Here, when the steering controller 11 inputs a steering command for correction from a lane keeping assist controller 15 described later, the steering command is added (added) to the steering command value by adding the corrected steering command to the steering command value. Correct the value. Further, the steering controller 11 outputs a steering command value corresponding to the steering command from the lane keeping support controller 15 described later to the steering actuator 25.

Further, each of the front wheel 13 and the rear wheel 40 is provided with a brake unit. Each brake unit includes a brake disk 22 and a wheel cylinder 23 that frictionally clamps the brake disk 22 to supply a braking force (braking force) by supplying hydraulic pressure. A pressure control unit 24 is connected to each wheel cylinder 23 of these brake units, and the brake unit individually applies braking to each wheel by the hydraulic pressure supplied from the pressure control unit 24.
A lane keeping assist device is provided for the host vehicle having the above system configuration.

The configuration will be described next.
A monocular camera with an image processing function is installed in the vehicle. This monocular camera with an image processing function is an external environment recognition means 16 for detecting the position of the host vehicle. A monocular camera with an image processing function images a road surface ahead of the host vehicle. The state of the road surface is determined from the captured camera image, and a signal related to the position of the host vehicle in the travel lane on which the host vehicle travels is output to the lane keeping support controller 15. The signal related to the position of the host vehicle in the travel lane is information regarding the yaw angle θ, which is an angular deviation in the traveling direction of the host vehicle with respect to the travel lane, the lateral displacement X from the center of the travel lane, and the curvature ρ of the travel lane.

Further, a direction indicating switch 17 is provided. A signal from the direction indicating switch 17 is output to the lane keeping support controller 15 as determination information as to whether or not the driver changes the traveling lane.
Further, the lane keeping assist controller 15 receives signals from the steering controller 11 such as the current steering state and tire steering state.
The lane keeping support controller 15 calculates a control amount for keeping the host vehicle in the traveling lane based on the input signal, and outputs it to at least the steering controller 11.

Next, the processing of the lane keeping support controller 15 will be described with reference to FIG.
The lane keeping support controller 15 is repeatedly executed every predetermined sampling period.
First, when it operates, in step S100, various data from each sensor and the steering controller 11 are read. Each wheel speed Vw is read from the wheel speed sensors 18-21. Further, the steering angle δ, the steering angular velocity δ ′, and the direction indication switch 17 signals are read. From the camera controller of the external recognition means 16, the yaw angle θ of the host vehicle with respect to the traveling lane L of the host vehicle, the lateral displacement X from the center of the traveling lane L, and the curvature ρ of the traveling lane L are read. Here, the lateral displacement X from the traveling lane center Ls may be based on the center of gravity G of the host vehicle C as shown in FIGS. 3 and 4, for example. However, the center of gravity position G of the host vehicle C may not be used as a reference. For example, the lateral displacement X from the traveling lane center Ls may be obtained with reference to the center of the front end of the host vehicle C. That is, as shown in FIG. 4, the vehicle is displaced in the departure direction first from the front end portion of the host vehicle C according to the yaw angle θ, so the lateral displacement X is obtained with reference to that portion, and the lateral displacement deviation is reduced earlier. You may make it do.

Subsequently, in step S110, the left and right lateral displacement reference threshold values XLt and XRt are set based on the following formulas (1) and (2).
Here, as shown in FIG. 3, the lateral displacement reference threshold value XRt on the right side specifies the position of the lateral displacement reference position LXR that is a reference for the deviation of the lateral displacement X set for the right departure. The left lateral displacement reference threshold XLt specifies the position of the lateral displacement reference position LXL, which is a reference for the lateral displacement X deviation set for the left departure.
XRt = (Wlane / 2)-(Wcar / 2)
-Xoffset (1)
XLt = − ((Wlane / 2) − (Wcar / 2)
-Xoffset) (2)
Here, the lateral displacement X from the travel lane center Ls is positive when the host vehicle C is on the right side of the center with respect to the travel lane L, and is negative when the vehicle is located on the left side. For this reason, the right side lateral displacement reference threshold value XRt side is positive.

Further, as shown in FIG. 3, Wlane is the travel lane width, and Wcar is the vehicle width of the host vehicle C.
Xoffset is a margin for the position of the traveling lane edge Le (white line or shoulder). This margin allowance Xoffset may be changed according to the travel lane width Wlane or the vehicle speed. For example, the margin Xoffset is reduced as the travel lane width Wlane is narrower. Further, different margin allowances Xoffset may be used for the left and right lateral displacement reference positions LXL and LXR. The left and right lateral displacement reference positions LXL and LXR may be fixed values.

Subsequently, in step S120, a lateral displacement deviation ΔXR with respect to the right departure is calculated based on the following equation (3).
ΔXR = X−XRt (3)
However, when ΔXR ≦ 0, ΔXR = 0 is set (only positive values are taken).
From the above equation (3), the lateral displacement X and the lateral displacement deviation ΔXR with respect to the right departure are in the relationship shown in FIG.
That is, by using the above expression (3), when “X−XRt ≧ 0”, it is determined that the host vehicle C has moved out of the right lateral displacement reference position LXR with respect to the traveling lane center Ls. Since the host vehicle C is close to the right traveling lane edge Le side, the right lateral displacement reference position LXR as the lateral displacement reference position close to the host vehicle C is used as the lateral displacement deviation reference, and the right deviation is prevented. The lateral displacement deviation ΔXR is obtained.

Subsequently, in step S130, a lateral displacement deviation ΔXL with respect to the left departure is calculated based on the following equation (4).
ΔXL = X−XLt (4)
However, when ΔXL ≧ 0, ΔXL = 0 (only negative values are taken).
From the above equation, the lateral displacement X and the lateral displacement deviation ΔXL with respect to the left departure are in the relationship shown in FIG.
That is, by using the above equation (4), when “X−XLt ≦ 0”, it is determined that the host vehicle C has moved out of the left lateral displacement reference position LXL with respect to the traveling lane center Ls. Then, since the host vehicle C is close to the left lane edge Le side, the left lateral displacement reference position LXL as the lateral displacement reference position close to the own vehicle C is used as a reference for the lateral displacement deviation. The lateral displacement deviation ΔXL is obtained.

Subsequently, in step S140, the yaw angle deviation ΔθR with respect to the right departure is calculated based on the following equation (5). Here, the yaw angle θ of the host vehicle C with respect to the lane L is positive when the yaw angle θ is on the right side (the state shown in FIG. 4), and negative when the yaw angle θ is on the left side. .
ΔθR = θ (when θ> 0)
ΔθR = 0 (when θ ≤ 0)
... (5)
According to the above equation (5), the yaw angle θ and the yaw angle deviation ΔθR set only for the right departure are in the relationship shown in FIG.

Subsequently, in step S150, the yaw angle deviation ΔθL with respect to the left departure is calculated based on the following equation (6).
ΔθL = θ (when θ <0)
ΔθL = 0 (when θ ≧ 0)
... (6)
According to the above equation (6), the yaw angle θ and the yaw angle deviation ΔθL set only for the left deviation have the relationship shown in FIG.

Subsequently, in step S160, the travel lane edge control feedback correction gain KρL_R for the right departure and the travel lane for the left departure according to the curve direction, the curvature ρ, and the yaw angle θ (deviation) of the travel lane L. End control feedback correction gains KρL_L are respectively obtained.
That is, it is divided into three types according to the direction of the curvature ρ (curve direction of the driving lane L), and using the individual maps as described below, the driving lane edge control feedback correction gain KρL_R for the right departure, and A lane edge control feedback correction gain KρL_L for the left departure is set.

When it is determined that the curvature ρ <0 (right curve):
KρL_R: Read from the curve IN side correction gain map as shown in FIG.
KρL_L: Read from the curve OUT side correction gain map as shown in FIG.
When it is determined that the curvature ρ> 0 (left curve), KρL_R: read from the curve OUT side correction gain map as shown in FIG.
KρL_L: Read from the curve IN side correction gain map as shown in FIG.
When it is determined that the curvature ρ = 0 (straight road) KρL_R = 1.0 (no correction)
KρL_L = 1.0 (no correction)
Here, the curvature ρ of the traveling lane L is the reciprocal of the turning radius, ρ = 0 on a straight road, and the absolute value of the curvature ρ increases as the curve becomes tighter (the turning radius becomes smaller). The left curve is positive and the right curve is negative.

  As shown in FIG. 7, the curve IN-side correction gain map is a map in which the correction gain decreases as the absolute value of the curvature ρ increases when the absolute value of the curvature ρ exceeds a predetermined value. And the gain of control with respect to the travel lane edge Le located inside the curved road among the left and right travel lane edges Le is corrected so as to decrease in accordance with the increase in the absolute value of the curvature ρ.

  Further, as shown in FIG. 8, the curve OUT side correction gain map is a map in which the correction gain increases as the absolute value of the curvature ρ increases as the absolute value of the curvature ρ increases to a predetermined value or more. And the gain of control with respect to the traveling lane edge Le located outside the curved road among the left and right traveling lane edges Le is corrected so as to increase in accordance with the increase in the absolute value of the curvature ρ.

Subsequently, in step S170, depending on the curve direction of the travel lane L, the curvature ρ, and the lateral displacement X (deviation), the travel lane center control feedback correction gain KρY_R for the right departure, the travel lane center control for the left departure. A feedback correction gain KρY_L is obtained.
That is, it is divided into three types according to the direction of the curvature ρ (curve direction of the travel lane L), and using the map as follows, the travel lane center control feedback correction gain KρY_R for the right departure and the left departure A lane center control feedback correction gain KρY_L is set.

When it is determined that the curvature ρ <0 (right curve) KρY_R: Read from a curve IN side correction gain map as shown in FIG.
KρY_L: Read from the curve OUT side correction gain map as shown in FIG.
When it is determined that the curvature ρ> 0 (left curve) KρY_R: Read from a curve OUT side correction gain map as shown in FIG.
KρY_L: Read from the curve IN side correction gain map as shown in FIG.
When it is determined that the curvature ρ = 0 (straight road) KρY_R = 1.0 (no correction)
KρY_L = 1.0 (no correction)

  Here, the curve IN side correction gain map and the curve OUT side correction gain map use the same trend map when obtaining the driving lane edge portion control feedback correction gain and when obtaining the driving lane center portion control feedback correction gain. doing. However, different maps are used for determining the traveling lane edge control feedback correction gains KρL_R and KρL_L and for determining the traveling lane center portion control feedback correction gains KρY_R and KρY_L. That is, the curve IN-side correction gain map and the curve OUT-side correction gain map used when determining the lane edge control feedback correction gains KρL_R and KρL_L set a larger gradient with respect to the change in the absolute value of the curvature ρ. This is because the amount of correction with respect to the curvature ρ on the inner side and the outer side of the curve is increased on the traveling lane end Le side, and the sensitivity is increased accordingly.

Subsequently, in step S180, based on the following formulas (7) and (8), the target turning angle φL_Rt by the traveling lane edge control for the right departure and the target turning angle φL_Lt by the traveling lane edge control for the left departure. Is calculated.
φL_Rt = − (((Kc_L1 × Kv_L1 × ΔXR)
+ (Kc_L2 × Kv_L2 × θ)
+ (Kc_L3 × Kv_L3 × ρ))
× KρL_R) (7)
φL_Lt = − (((Kc_L1 × Kv_L1 × ΔXL)
+ (Kc_L2 × Kv_L2 × θ)
+ (Kc_L3 × Kv_L3 × ρ))
× KρL_R) (8)

Here, Kc_L1, Kc_L2, and Kc_L3 are feedback gains determined by vehicle specifications.
Kv_L1, Kv_L2, and Kv_L3 are correction gains corresponding to the vehicle speed. For example, Kv_L1, Kv_L2, and Kv_L3 increase according to the vehicle speed.
Here, the two items and the three items in the equations (7) and (8) are correction terms (convergence terms) for the lateral displacement deviation. For this reason, the correction gains Kc_L2 and Kc_L3 are set smaller than the correction gain Kc_L1. Similarly, the correction gains Kc_L2 and Kc_L3 are set smaller than the correction gain Kv_L1.

  That is, the target turning angles φL_Lt and φL_Rt based on the lane edge control for the right departure or left departure are used to obtain a control amount that reduces the lateral displacement deviation from the lateral displacement reference positions LXL and LXR. At that time, the control amount is corrected by the yaw angle θ and the road curvature ρ of the host vehicle C. At this time, the yaw angle θ of the host vehicle C in the second term of the above formula acts as a feedback control amount for the lateral speed. For this reason, the yaw angle θ is used as the yaw angle θ of the vehicle C in the second term without using the yaw angle deviation ΔθR or ΔθL.

  From the above, when the final final target turning angles φft and φrt are calculated as in step S200 described later, the target turning angle φL_Rt by the driving lane edge control for the right departure and the driving lane edge control for the left departure are calculated. Is calculated as the sum of the target turning angle φL_Lt. That is, the sum of the target turning angle φL_Rt and the target turning angle φL_Lt is set as the target turning angle for the traveling lane edge control.

  At this time, when the host vehicle C is positioned between the left and right lateral displacement reference positions LXL and LXR on the traveling lane center Ls side, both the lateral displacement deviations ΔXR and ΔXL are 0 as shown in FIG. Become. That is, the values of the target turning angle φL_Rt and the target turning angle φL_Lt are small values. As a result, the target turning angle for the traveling lane edge control becomes small, and the target turning angles φY_Lt and φY_Rt for the traveling lane center control described later are dominant.

  Further, when the host vehicle C is located outside the left and right lateral displacement reference positions LXL and LXR on the traveling lane center Ls side, only one value of the lateral displacement deviations ΔXR and ΔXL is obtained as shown in FIG. 0. That is, of the target turning angle φL_Rt and the target turning angle φL_Lt, one of the target turning angle φL_Lt or φL_Rt for controlling the traveling lane edge on the side far from the own vehicle C becomes smaller and closer to the own vehicle C. The other of the target turning angle φL_Lt or φL_Rt for controlling the traveling lane edge is dominant as the target turning angle for the traveling lane edge control.

Then, the yaw angle θ of the host vehicle C with respect to the traveling lane L is provided as a differential term (lateral velocity) with respect to the lateral displacement X in the second term and fed back as it is, and further the third term as a correction term for the road curvature ρ. Provide and control by feedback. As a result, according to the first term, the second term and the third term are appropriately retained within the travel lane L while eliminating the movement to control the position of the host vehicle C outside the travel lane L with respect to the lateral displacement reference position. By providing this, it is possible to reduce the feeling of the vehicle C repelling from the traveling lane edge Le. That is, by providing the second term (differential value of lateral displacement) and the third term (convergence term with respect to the road surface curve) as convergence terms, convergence to the lateral displacement reference position is improved.
Further, the control gain is corrected by multiplying the lane edge control feedback correction gains KρL_R and KρL_L. That is, by correcting according to the curve direction, the curvature ρ, and the lateral position of the traveling lane L, it is possible to appropriately control the vehicle on the curved road without feeling uncomfortable.
The third term may be zero.

Subsequently, in step S190, based on the following formulas (9) and (10), the target turning angle φY_Rt by the driving lane center control for the right departure and the target turning angle φY_Lt by the driving lane control for the left departure are calculated. To do.
φY_Rt = − (Kc_Y × Ky_R × Kv_Y × KρY_R × ΔθR)
... (9)
φY_Lt = − (Kc_Y × Ky_L × Kv_Y × KρY_L × ΔθL)
(10)
Here, Kc_Y is a feedback gain determined by vehicle specifications. Kv_Y is a correction gain according to the vehicle speed. For example, Kv_Y is set to a larger value as the vehicle speed is higher.

Ky_R and Ky_L are feedback gains that are individually set according to the lateral displacement X with respect to the travel lane L as shown in FIG.
That is, the target turning angle φY_Rt based on the driving lane center control for the right departure is a case where the traveling direction of the host vehicle C is facing the right side. For this reason, the feedback gain Ky_R with respect to the right departure is set so as to increase as it approaches the right travel lane end Le with respect to the left travel lane end Le.

  In addition, the target turning angle φY_Lt by the driving lane center control for the left departure is a case where the traveling direction of the host vehicle C is directed to the left side. For this reason, the feedback gain Ky_L with respect to the left departure is set so as to increase as it approaches the left traveling lane end Le with reference to the right traveling lane end Le side. Note that the target turning angles φY_Rt and φY_Lt are positive for rightward turning and negative for leftward turning.

  Here, as in step S200, which will be described later, the final target turning angle for the driving lane center control is set to the target turning angle φY_Rt by the driving lane center control for the right departure and the target by the driving lane control for the left departure. Calculated as the sum of the turning angle φY_Lt. At this time, when the yaw angle θ is on the right side, ΔθL = 0 as shown in FIG. 6, so the target turning angle φY_Lt for the left departure is 0. That is, only the target turning angle φY_Rt for the right departure is adopted. Similarly, when the yaw angle θ is on the left side, ΔθR = 0 as shown in FIG. 6, so the target turning angle φY_Rt for the right departure is 0. That is, only the target turning angle φY_Lt for the left departure is adopted.

  At this time, as described above, the control gains Ky_R and Ky_L become closer to the traveling lane end Le with reference to the traveling lane end Le on the direction of the yaw angle θ of the host vehicle C as shown in FIG. It is set to be larger. Therefore, when the yaw angle θ is generated on the departure side, the control is performed with a large amount of control so as to actively prevent the departure. In addition, when the yaw angle θ is generated toward the departure avoidance side, the direction of the traveling direction of the host vehicle C is gently changed in the direction along the traveling lane L without causing a sense of incongruity by reducing the control amount. Can be matched.

Further, the control gains Ky_R and Ky_L are set so as to increase toward the traveling lane end Le with reference to one traveling lane end Le. As a result, even when the host vehicle C travels across the travel lane center Ls, the control amount continuously changes, and it is possible to suppress a sense of discomfort when straddling the travel lane center Ls.
Further, when the vehicle is displaced inward or outward with respect to the running lane center Ls on the curve road, by correcting according to the curve direction and the curvature ρ of the running lane L as calculated in step S170, Even on curved roads, the control can be performed appropriately without feeling uncomfortable.

Subsequently, in step S195, rear wheel steering gains Kry and Krl, and front wheel steering gains Kfy and Kfl are calculated.
First, rear wheel steering gains Kry and Krl are calculated based on the following equations.
Kry = Ksy × Key × Kvy × Kdy (11)
Krl = Ksl × Kel × Kvl × Kdl (12)
Here, Ksy and Ksl are gains determined by vehicle specifications. Here, the gain Ksy is set to be a negative value. The gain Ksl is set to be a positive value.

The gains Key and Kel are gains that change according to the traveling environment around the vehicle, such as the lateral position of the host vehicle in the lane and the curvature of the traveling lane.
For example, Key and Kel are set as shown in FIGS. That is, on the vehicle center side, Key is small and Kel is large. Conversely, on the lane edge side, Key is large and Kel is small. In this case, in the rear wheel steering, the lateral position control is dominant on the lane center side, and the yaw angle control is dominant on the lane end side.

The gains Kvl and Kvy are gains that change depending on the vehicle speed.
For example, Kvl and Kvy are determined as shown in FIGS. That is, the gain Kvy is set to a high gain at a low vehicle speed and is set to a normal value at a medium vehicle speed. At higher vehicle speeds, the gain is set low. Conversely, the gain Kvl tends to be reversed. As a result, yaw angle control becomes dominant on the low speed side, and lateral position control becomes dominant on the high speed side.

Kdl and Kdy are driver steering intention estimation values estimated from torque sensors or steering angle sensors provided on the steering.
For example, Kdl and Kdy are determined as shown in FIGS. Kdl and Kdy are large values when the steering torque is small, and are small when the steering torque is greater than or equal to a predetermined value.
Here, Kry is set to be a negative value, and Krl is set to be a positive value.
Further, the gains Key, Kvy, Kdy, Ksl, Kel, Kvl, and Kdl need not all be used, and may be appropriately selected or combined. Moreover, it is not necessary to use all of these gains.

Next, based on the following formula, front wheel steering gains Kfy and Kfl are calculated according to the rear wheel steering gains Kry and Krl.
Kfy = 1 + Kry (13)
Kfl = 1 + Krl (14)
Here, Kry is negative and Krl is positive. For this reason, the larger the absolute value of Kry, the smaller the value of Kfy. On the other hand, Kfl increases as Krl increases.
The front wheel steering gains Kfy and Kfl may be set as follows. That is, the rear wheel turning amount may not be taken into consideration.
Kfy = 1
Kfl = 1
Subsequently, in step S200, the final target turning angle φft of the front wheels and the final target turning angle φrt of the rear wheels for lane keeping support are calculated.

In this embodiment, the final target turning angle φft of the front wheels and the final target turning angle φrt of the rear wheels are set to the left and right target turning angles φL_Lt by the lane edge control calculated in step S180, It is calculated as the sum of φL_Rt and the left and right target turning angles φY_Lt and φY_Rt by the driving lane center control calculated in step S190.
φft = (Kfl × α_R × φL_Rt + Kfy × β_R × φY_Rt)
+ (Kfl × α_L × φL_Lt + Kfy × β_L × φY_Lt)

φrt = (Krl × α_R × φL_Rt + Kry × β_R × φY_Rt)
+ (Krl × α_L × φL_Lt + Kry × β_L × φY_Lt)

Here, α_R and β_R are weighting coefficients for the driving lane edge control for the right departure and the driving lane center control, respectively. Α_L and β_L are weighting coefficients for the driving lane edge control for the left departure and the driving lane center control, respectively.
The weighting coefficients α_R, α_L, β_R, and β_L have the relationship shown in FIG. 10 so that the relative sizes of β_R and β_L with respect to α_R and α_L change according to the lateral position of the host vehicle C. It has become.
Moreover, it has the relationship of the following formula.
α_R + β_R = 1.0
α_L + β_L = 1.0

This weighting coefficient will be described.
As shown in the equations (7) and (8), there is feedback of the yaw angle element (lateral velocity) as the second term of the target turning angles φL_Rt and φL_Lt by the traveling lane edge control. This feedback is set as a differential term of the lateral displacement element for reducing the feeling of being repelled from the traveling lane edge Le. For this reason, it is possible to improve the convergence to the lateral displacement reference position together with the feedback of the lateral displacement element.

On the other hand, as shown in the equations (9) and (10), the feedback of the yaw angle θ element in the traveling lane center control is set for the purpose of matching the traveling direction of the host vehicle C with the traveling lane L. .
Therefore, for example, when the left end portion of the travel lane L has a yaw angle θ toward the left side (deviation side), in addition to the lateral displacement feedback element in the travel lane end control, the yaw angle in the travel lane center control If feedback is performed, there is a risk of excessive control. When the yaw angle θ is on the right side (deviation avoidance side) at the left end of the lane L, the yaw angle feedback in the lane center control is set weak and the convergence to the lateral displacement reference position is poor. There is a possibility that a feeling of being repelled from the traveling lane end Le will occur.

  For this reason, in this embodiment, for example, as shown in FIG. 16, the weight on the traveling lane edge control side is increased as approaching the traveling lane edge Le side from the lateral displacement reference threshold. On the other hand, as the vehicle approaches the lane center Ls, the weight on the lane center control side is increased. In this way, these weights are set according to the lateral position of the host vehicle C with respect to the travel lane L. By setting in this way, in the driving lane center Ls, free line removal without a sense of restraint is realized, and at the driving lane end Le, the driving lane end Le is appropriately retained in the driving lane L and from the driving lane end Le. The feeling of being repelled can be reduced.

By performing weighting as described above, there is an overlapping control region in which both lateral displacement feedback and yaw angle feedback control are performed on the lane edge Le side of the lateral displacement reference threshold.
Subsequently, in step S210, the driver's intention to change the driving lane is determined. More specifically, the driver's intention to change the driving lane L is determined based on the direction switch signal obtained in step S100 and the traveling direction of the host vehicle C.

  That is, when the direction indicated by the direction switch signal (the blinker lighting side) and the traveling direction of the host vehicle C are the same direction, it is determined that the driver is intentionally changing the driving lane L. In this case, the process returns without correcting the turning angle in step S220. Note that when the steering wheel 12 is steered in the same direction as the direction switch signal (the blinker lighting side), it may be determined that the driver is intentionally changing the driving lane L.

Subsequently, in step S220, the corrected turning angle command value of the final target turning angle φft on the front wheel side calculated in step S200 is output to the steering controller 11. Further, the calculated turning angle command value of the final target turning angle φrt on the rear wheel side is output to the steering controller 11.
Here, in the steering controller 11, as described above, when the corrected turning angle command value of the final target turning angle φft is input from the lane keeping support controller 15, the target turning angle calculated in accordance with the driver's steering operation. Is added with the final target turning angle φft on the front wheel side to obtain the final target turning angle, and the front wheel turning actuator 5 is driven so that the turning angle corresponding to the target turning angle is obtained. To do. Further, when the steering controller 11 receives the turning angle command value of the final target turning angle φrt on the rear wheel side from the lane keeping assist controller 15, the steering controller 11 turns according to the turning angle command value of the final target turning angle φrt. The rear wheel steering actuator 25 is driven so that the steering angle is obtained.

Here, the host vehicle C of the steer-by-wire system is illustrated as the host vehicle C that provides the lane keeping assist device of the present embodiment. In the case of the own vehicle C equipped with a power steering system using electric or hydraulic pressure, the final target turning angles φft and φrt are converted into assist torque amount correction amounts and added to the assist torque to turn the vehicle. You may make it correct | amend a corner part.
Further, in the own vehicle C in which the steering shaft can be rotationally displaced to change the turning angle, the rotational displacement amount may be corrected by the final target turning angle φft.

Here, the left and right lateral displacement reference positions LXL and LXR constitute both the lateral displacement threshold and the lateral displacement reference position. Steps S180 and S190 constitute control amount calculation means. Step S200 and the steering controller 11 constitute a traveling direction control means. Equations (4) and (5) correspond to a configuration in which the lateral displacement deviation is zero or the control gain is small when it is located between the left and right lateral displacement thresholds. Equations (9) and (10) constitute second control amount calculation means, third control amount calculation means, and fourth control amount calculation means, and the target turning angles φY_Rt and φY_Lt are the second control amounts, It becomes the third control amount and the fourth control amount. Equations (7) and (8) constitute the first control amount calculation means, and the target turning angles φL_Rt and φL_Lt are the first control amounts. Further, the expression (11) constitutes a final control amount calculation unit. Further, the final target turning angles φft and φrt constitute the final control amount. Further, the correction gains KρL_R, KρL_L, KρY_R, and KρY_L constitute a curve road correction unit. The left and right traveling lane edge Le constitutes the lateral edge reference position. The weighting coefficients α_R and α_L constitute the second weighting coefficient and the third weighting coefficient. The weighting coefficients β_R and β_L constitute the first weighting coefficient. The yaw angle deviation ΔθR or ΔθL constitutes the angle deviation. The second term of the equations (7) and (8) constitutes the fifth control amount.
Steps S195 and S200 constitute turning direction adjusting means.

(Operation)
“Between left and right lateral displacement reference positions LXL and LXR”
First, a case where the host vehicle C is traveling between the left and right lateral displacement reference positions LXL and LXR will be described.
In this case, both ΔXR and ΔXL are zero. For this reason, the first terms of the left and right target turning angles φL_Rt and φL_Lt by the driving lane edge control shown in the equations (7) and (8) are zero. That is, the left and right target turning angles φL_Rt and φL_Lt by the driving lane edge control are small values.
Here, the two terms of the left and right target turning angles φL_Rt and φL_Lt by the traveling lane edge control are control amounts that make the lateral speed of the host vehicle C zero. Further, the two terms of the left and right target turning angles φL_Rt and φL_Lt by the traveling lane edge control take a value of zero if the vehicle is a straight road.

Further, when the host vehicle C is traveling between the left and right lateral displacement reference positions LXL and LXR, the weighting coefficients α_R and α_L for the left and right target turning angles φL_Rt and φL_Lt by the traveling lane edge control are shown in FIG. A small value such as 10 is set.
Accordingly, when the host vehicle C is traveling between the left and right lateral displacement reference positions LXL and LXR, the target turning angles φY_Rt and φY_Lt based on the traveling lane center portion control are dominant. In particular, as shown in FIG. 16, rather than the weighting coefficients α_R and α_L for the left and right target turning angles φL_Rt and φL_Lt by the traveling lane edge control, β_R and β_L of the target turning angles φY_Rt and φY_Lt by the traveling lane center control Is set to be larger. For this reason as well, when the host vehicle C is traveling between the left and right lateral displacement reference positions LXL and LXR, the target turning angles φY_Rt and φY_Lt based on the traveling lane center control are dominant.

For this reason, when the own vehicle C is traveling between the left and right lateral displacement reference positions LXL and LXR, the angle deviation is controlled so that the own vehicle C travels as intended by the driver. Along the line, the vehicle travels in a direction parallel to the travel lane L.
From the above, when the host vehicle C is located on the traveling lane center Ls side, the angle deviation is controlled to be small. Also, there is little or no feedback for lateral displacement. That is, there is no or little control intervention to return to the traveling lane center Ls side. As a result, the host vehicle C travels along the travel lane L according to the travel line intended by the driver.

Further, in the present embodiment, the steering of the rear wheels is also controlled in conjunction with the control of the steering of the front wheels so that the angle deviation becomes small. At this time, since the target turning angles φY_Rt and φY_Lt controlled by the traveling lane center control are dominant, the influence of the gains Kfl and Krl is small.
The gain Kry is a negative value, whereas the gain Kfy is a positive value. For this reason, the rear wheels are steered in a direction opposite to the steered direction of the front wheels. As a result, as shown in FIG. 17, the followability of the vehicle motion with respect to the turning control and the convergence of the control are improved as compared with the case where the turning control is performed only with the front wheels.

Further, the gain Kfy is relatively decreased as the gain Kry is increased. As a result, the turning performance given to the driver is increased compared with the steering control of only the front wheels, while the yaw rate generated by the vehicle is equivalent to the state of only the front wheels.
At this time, when calculating the target turning angles φY_Rt and φY_Lt, as shown in the equations (9) and (10), the control gains are multiplied by Ky_R and Ky_L for correction. The control gains Ky_R and Ky_L increase as the distance of the host vehicle C approaches the traveling direction end on the traveling direction side of the host vehicle C, and the target turning angles φY_Rt and φY_Lt become large values.

For this reason, when the traveling direction of the host vehicle C is on the departure side, the target turning angles φY_Rt and φY_Lt are increased and the departure prevention effect is increased. Further, when the traveling direction of the host vehicle C is on the departure avoidance side, the target turning angles φY_Rt and φY_Lt are reduced to reduce a sense of discomfort without excessive control.
For example, when the traveling direction of the host vehicle C has a yaw angle θ on the right side with respect to the travel lane L, the host vehicle C is laterally displaced X to the right with respect to the travel lane center Ls and is positioned (deviation side). The target turning angle φY_Rt increases as the distance increases. That is, the deviation avoidance effect is increased. On the other hand, the target turning angle φY_Rt becomes smaller as the host vehicle C is displaced laterally X to the left with respect to the traveling lane center Ls (deviation avoidance side).

  Further, the control gains Ky_R and Ky_L are changed in accordance with the distance from the traveling direction end on the traveling direction side of the host vehicle C. For this reason, when the traveling direction of the host vehicle C is tilted to the right with respect to the travel lane L, that is, when the yaw angle θ is on the right side, the host vehicle C moves from the left to the right with respect to the travel lane center Ls. Even when the vehicle travels across the center Ls, it is difficult for the driver to feel uncomfortable.

“Transition period to enter outside the lateral displacement reference positions LXL and LXR”
Next, a description will be given of a case where the host vehicle C moves outward from the left and right lateral displacement reference positions LXL and LXR from the travel lane center Ls side.
Here, the area on the side of the traveling lane from the lateral displacement reference positions LXL and LXR is referred to as a departure area.
As described above, when the host vehicle C is traveling between the left and right lateral displacement reference positions LXL and LXR, control is performed such that the angular deviation is reduced. For this reason, when the own vehicle enters the departure area, the yaw angle (angle deviation) in the departure direction of the own vehicle is reduced.
That is, in the process in which the host vehicle enters the departure area, the control by the second control amount that reduces the angular deviation toward the departure side is the preliminary control for reducing the first control amount that reduces the lateral displacement deviation. Acts as

"When located in the departure area"
Next, a case where the host vehicle C is positioned outside the left and right lateral displacement reference positions LXL and LXR with respect to the travel lane center Ls (when located in the departure area) will be described.
In this case, the deviation from the lateral displacement reference position on the side closer to the own vehicle C is reduced by the target turning angles φL_Rt and φL_Lt by the driving lane edge control shown by the equations (7) and (8). Control intervenes. That is, a control for returning to the left and right lateral displacement reference positions LXL and LXR with respect to the travel lane center Ls, that is, to the travel lane center Ls side, is involved. As a result, the movement to control the position of the host vehicle C to the outside of the traveling lane L can be eliminated, and the vehicle can be appropriately kept in the traveling lane L.

  At this time, the gains Kfl and Krl have the same sign. Therefore, the control for reducing the deviation from the lateral displacement reference position on the side closer to the own vehicle C is performed by turning the rear wheels in the same phase with respect to the turning direction of the front wheels as shown in FIG. To do. For this reason, the yaw angle to the inside of the lane, which is generated in order to reduce the deviation from the lateral displacement reference position, is reduced compared to the steering control using only the front wheels. As a result, control stability and convergence are improved.

Furthermore, the amount of steering of the front wheels increases as the amount of steering of the rear wheels for reducing the deviation from the lateral displacement reference position increases. As a result, the turning feeling given to the driver can be made smaller than that of only the front wheel steering while the yaw rate generated in the vehicle is equivalent to the state of only the front wheels.
At this time, the control for reducing the angle deviation by the target turning angles φY_Rt and φY_Lt by the driving lane center control is also intervening.

  For this reason, when there is an angle deviation (yaw angle θ) on the departure side (the end side of the travel lane L close to the host vehicle C) as in the lower part of FIG. 19, the angle deviation is eliminated. Along with the control amount, a control amount in the direction to cancel the lateral displacement is generated in the same direction. As a result, the amount of control to the departure avoidance side is increased, and departure can be prevented more effectively. At this time, as described above, the control gains Ky_R and Ky_R of the yaw angle feedback are large values. That is, since the control amount for eliminating the angle deviation is large, the effect is great.

  At this time, the turning amount of the rear wheel and the phase of turning with respect to the front wheel are determined by the amount of the control amount that eliminates the angular deviation and the control amount that eliminates the lateral displacement. For example, the rear wheels are steered with respect to the steered direction of the front wheels to the more dominant control side of the control amount that eliminates the angular deviation and the control amount that eliminates the lateral displacement.

  Further, when the angle deviation (yaw angle θ) is on the departure avoidance side (the direction away from the end portion side of the travel lane L close to the own vehicle C) as in the upper part of FIG. 19, the angle deviation is eliminated. Depending on the control amount toward the departure side, the control amount in the direction to cancel the lateral displacement is reduced or eliminated. As a result, a sense of incongruity as if the control is intervening on the departure side can be reduced. At this time, as described above, the control gains Ky_R and Ky_R of the yaw angle feedback are small values. That is, since the control amount for eliminating the angle deviation is small, the effect of reducing the sense of incongruity is great.

Further, as shown by the equations (7) and (8), the target turning angles φL_Rt and φL_Lt by the driving lane edge control are controlled by the second term to reduce the lateral speed, and the third term is the road. By the control amount with the curvature ρ taken into account, the convergence in the direction along the lateral displacement reference position is improved, and the feeling of being repelled from the traveling lane edge Le can be reduced.
Further, by correcting the control amount with the control feedback correction gains KρL_R, KρL_L, KρY_R, and KρY_L, the target turning angles φft and φrt are changed between the curve inside and the curve outside on the curved road.

That is, as shown in the lower part of FIG. 20, when the host vehicle C is positioned inside the curve with respect to the travel lane center Ls, the control feedback correction gain increases as the curvature of the travel lane L increases, that is, the curve becomes tighter. KρL_R, KρL_L, KρY_R, and KρY_L are reduced. That is, the target turning angle is corrected to be small. This prevents excessive control and reduces the uncomfortable feeling to the driver.
On the other hand, when the host vehicle C is positioned outside the curve with respect to the travel lane center Ls as in the upper part of FIG. 20, the control feedback correction gain KρL_R increases as the curvature of the travel lane L increases, that is, the curve becomes tighter. , KρL_L, KρY_R, and KρY_L are increased. That is, the target turning angle is greatly corrected. This increases the driving lane departure effect.

"Complex action when deviating beyond the lateral displacement threshold"
FIG. 21 shows the trajectory of the host vehicle when the vehicle deviates outside the lateral displacement threshold.
As described above, the yaw angle control can reduce the yaw angle (entrance angle) in the departure direction of the host vehicle when the host vehicle enters the departure region.
For this reason, the amount of deviation to the outside of the lateral displacement threshold after the host vehicle enters the departure area is reduced. As a result, the control amount (lateral position control) for reducing the lateral displacement deviation is reduced. Since this control amount is small, the feeling of being repelled from the end of the traveling lane is reduced accordingly. It also leads to an improvement in deviation avoidance ability.

Further, in a scene in which the angle deviation of the host vehicle is in the departure avoidance direction and approaches the lateral displacement threshold value, the feeling of rebound is reduced as described above due to the synergistic effect of the yaw angle and the lateral position control.
At this time, the rear wheel reverse phase steering is used for yaw angle control, and the rear wheel in-phase steering is combined with lateral position control. Since the vehicle is guided inward and the control followability with respect to the generation of the yaw angle is high on the lane center side, the vehicle fits in the lane center. That is, control convergence is good.

(Effect of this embodiment)
(1) The second control amount is calculated by the second control amount calculation means. That is, yaw angle feedback control, which is control for reducing the angle deviation, is performed within the left and right lateral displacement thresholds of the host vehicle C, that is, on the side of the traveling lane center portion relative to the lateral displacement threshold. Thus, the traveling direction of the vehicle is controlled in the direction along the traveling lane L on the traveling lane center Ls side. As a result, on the traveling lane center Ls side, a free line can be taken without a sense of restraint.

(2) The first control amount is calculated by the first control amount calculation means. That is, lateral position feedback control (lateral position control), which is a control for reducing the lateral displacement deviation, is performed outside the lateral displacement threshold value of the left and right sides of the host vehicle C, that is, on the side of the traveling lane edge side from the lateral displacement threshold value.
As a result, when the host vehicle C enters a departure region outside the lateral displacement threshold, an effect of returning to the lateral displacement threshold is generated.

(3) At this time, when the own vehicle C is within the lateral displacement threshold value on the left and right, the yaw angle feedback control is performed by the second control amount. As a result, the own vehicle exceeds the lateral displacement threshold value from the center of the traveling lane. When entering the departure area, the entry angle can be kept small. Thus, the yaw angle feedback control using the second control amount has an effect as a preliminary control for preventing departure.
In other words, when the approach angle is small, the deviation amount to the outside of the lateral displacement threshold value after the following own vehicle enters the deviation region becomes small. When the host vehicle enters the departure area, the first control amount used in the feedback control for reducing the lateral displacement deviation is reduced.
As a result, the feeling of being repelled from the traveling lane end side can be reduced. That is, it is possible to reduce the sense of restraint felt by the occupant.

(4) The steered direction adjusting means sets the phase of the steered direction of the rear wheels relative to the steered direction of the front wheels in accordance with at least one of the first controlled variable and the second controlled variable. Control.
This makes it possible to improve the convergence of the control compared to the steering control of only the front wheels by controlling the phase of the steering direction of the rear wheels relative to the front wheels according to the angle deviation and lateral deviation. It becomes.
(5) The turning direction adjusting means turns the rear wheels in the same phase as the turning direction of the front wheels according to the first control amount.
By turning the rear wheels to the same phase with respect to the lateral position control, the yaw angle to the inside of the lane generated by the lateral position control is reduced as compared with the steering control with only the front wheels. As a result, control stability and convergence are improved.

(6) The front wheel turning amount is increased in accordance with the increase in the rear wheel turning amount.
While making the yaw rate generated in the vehicle equal to the state of only the front wheels, the turning feeling given to the driver is made smaller than that of only the front wheel steering.
That is, the rear wheels are steered to the same phase with respect to the lateral position control. That is, the front wheel steering control amount increases and the rear wheels steer in the same phase. Thereby, the turning feeling of the vehicle can be reduced as compared with the lateral position control using only the front wheels. As a result, the wall feeling can be reduced and the vehicle can be naturally guided into the lane in the case of a lane departure in which the lateral position is greatly generated.

(7) The turning direction adjusting means turns the rear wheels in a direction opposite to the turning direction of the front wheels according to the second control amount.
The rear wheels are steered to the opposite phase to the yaw angle control. As a result, the tracking and control convergence of the vehicle motion with respect to the steering control is improved as compared with the steering control using only the front wheels.
(8) The steering amount of the front wheel is decreased in accordance with the increase in the steering amount of the rear wheel.
As a result, the turning feeling given to the driver can be increased as compared with the front wheel steering alone, while the yaw rate generated in the vehicle is equivalent to the state of only the front wheels.
That is, the rear wheels are steered to the opposite phase with respect to the yaw angle control. That is, the rear wheels are set to reverse phase steering while reducing the front wheel steering control amount. As a result, the turning feeling of the vehicle can be increased as compared with the yaw angle control using only the front wheels. As a result, it is possible to increase the sense of wall with respect to the lane departure in which the yaw angle is large and to convey the departure risk from the vehicle behavior to the driver.

(9) The turning direction adjusting means corrects the turning of the rear wheel according to the traveling environment around the vehicle.
As a result, the turning of the rear wheels can be corrected based on the position in the width direction of the road of the vehicle, the curved road, and the like.
For example, the control is performed so that the generation of the yaw rate is small near the center of the lane, and the control is performed so that the generation of the yaw rate is large near the end of the lane. As a result, information from the vehicle motion corresponding to the lane departure risk can be given to the driver.

(10) The turning direction adjusting means corrects the turning of the rear wheel according to the vehicle speed.
Thereby, for example, when the vehicle speed is higher than when the vehicle speed is low, it is possible to suppress the steering of the rear wheel.
For example, the gain is set high at low vehicle speeds, and the gain is set as normal at medium vehicle speeds. At higher vehicle speeds, the gain is set low. This makes it possible to adjust the amount of yaw rate (information from the vehicle behavior to the driver) with respect to the inside of the lane without reducing the amount by which the vehicle is returned to the lane center side by the control.

(11) The turning direction adjusting means corrects the turning of the rear wheel in accordance with the estimated value of the driver steering intention, and suppresses the turning of the rear wheel when the steering intention is estimated.
Control is performed so that the yaw rate does not occur during driver steering, and control is performed with the usual yaw gain when the driver is not steering. As a result, it is possible to prevent an abrupt yaw rate from occurring when, for example, the driver's steering direction coincides with the control direction. That is, a yaw rate as intended by the driver can be generated.

(12) The traveling direction control means controls the control amounts of both the first control amount and the second control amount in at least a partial region (overlapping control region) in the range on the lane edge side from the lateral direction displacement threshold. Control based on.
Thereby, if the own vehicle C is outside the lateral displacement threshold value on the left and right with respect to the traveling lane center Ls, feedback control of both the lateral displacement X and the yaw angle θ is performed. As a result, on the traveling lane end Le side, it is possible to perform the lane maintenance support in which the host vehicle C is appropriately kept in the traveling lane L and the feeling of being repelled from the traveling lane end Le is reduced.

That is, in the overlap control region on the lateral displacement threshold side in the departure region, both the control amount for the lateral position control and the control amount for the yaw angle control are used. As a synergistic effect using both control amounts, the following effects are obtained.
That is, when the vehicle advances with an angular deviation toward the departure side when entering the departure area, the direction of the control amount of the lateral position control and the direction of the control amount of the yaw angle control are the same direction (the departure avoidance direction). Control amount. As a result, the turning radius of the vehicle can be increased.

On the other hand, when the vehicle travels with an angular deviation toward the departure avoidance when heading for the lateral displacement threshold in the departure region, the direction of the control amount of the lateral position control and the control amount of the yaw angle control are opposite directions. It becomes the control amount. As a result, the trajectory of the vehicle gradually approaches the lateral displacement threshold, that is, the turning radius of the vehicle can be increased.
Thus, by adding the yaw angle control to the lateral position control, the turning radius of the vehicle for avoiding deviation can be increased. As a result, the acceleration / deceleration in the yaw direction can be reduced, and the rebound can be more reliably reduced.

(13) The traveling direction control means determines the final control amount from the sum of the value obtained by multiplying the first control amount by the first weighting factor and the value obtained by multiplying the second control amount by the second weighting factor. calculate. The first weighting coefficient and the second weighting coefficient are changed according to the lateral displacement deviation, and the larger the lateral displacement deviation, the larger the first weighting coefficient is set with respect to the second weighting coefficient.
That is, the final target turning angles φft and φrt are calculated from the target turning angles φL_Rt and φL_Lt based on the traveling lane edge control and the target turning angles φY_Rt and φY_Lt based on the traveling lane center control. At this time, the first weighting coefficient and the second weighting coefficient are changed according to the lateral displacement deviation, and the larger the lateral displacement deviation, the larger the second weighting coefficient is set with respect to the first weighting coefficient. That is, the weight of the driving lane edge portion control is increased as approaching the driving lane edge portion Le side. On the other hand, as the vehicle approaches the lane center Ls side, the weight of the lane center control is increased. In this way, these weights are set according to the lateral position of the host vehicle C with respect to the travel lane L.

As a result, on the traveling lane center Ls side, yaw angle feedback is dominant, and free line removal without a sense of restraint is realized. On the other hand, on the traveling lane end Le side, the feedback control by the lateral displacement X becomes dominant, and it is possible to appropriately stay in the traveling lane L and reduce the feeling of being repelled from the traveling lane end Le. .
In particular, in this embodiment, there is a yaw angle feedback as a convergence term (second term) in the target turning angles φL_Rt and φL_Lt by the driving lane edge control, but the yaw angle feedback becomes excessively controlled by the weighting. This can be reduced.

(14) The first control amount is corrected by the lateral displacement speed. That is, the control amount of the lateral displacement speed is added as the second term of the target turning angles φL_Rt, φL_Lt by the traveling lane edge control, that is, the feedback control control amount by the lateral displacement X.
As a result, the convergence with respect to the lateral displacement X with respect to the lateral displacement reference position is improved. As a result, the feeling of being repelled from the traveling lane end Le can be further reduced.
In addition, there is an effect of increasing the turning radius of the vehicle when turning from the departure side to the departure avoidance direction.

(15) Of the left and right traveling lane edge Le, the second control amount (third third) by the traveling lane center portion control according to the distance of the own vehicle C with respect to the traveling lane edge Le located on the traveling direction side of the own vehicle Control gain, the fourth control amount) is corrected. Then, target turning angles φY_Rt and φY_Rt by the control of the traveling lane center are obtained.
That is, the control gain of the yaw angle feedback control is changed according to the lateral position of the vehicle with respect to the traveling lane L. At this time, it correct | amends so that the said control gain may become large, so that the distance of the own vehicle C with respect to the said travel lane edge part Le is short.
For example, when the yaw angle θ is on the right side, the yaw angle θ increases from the left traveling lane end Le toward the right traveling lane end Le. Further, when the yaw angle θ is on the left side, the value is set to increase as it approaches the left travel lane end Le from the right travel lane end Le.

As a result, even if feedback control based on the lateral displacement X is not performed, by controlling the traveling direction (yaw angle θ) of the vehicle, there is no sense of incongruity (a sense of restraint) resulting from a deviation from the travel line intended by the driver. Lane maintenance support can be performed.
When the yaw angle θ is on the departure side, the control gain (control amount) is increased to ensure the departure prevention effect. On the other hand, when the yaw angle θ is on the departure avoidance side, the control gain (control amount) can be reduced, and the uncomfortable feeling of excessive control can be reduced.
At this time, even if the control gain (control amount) when the yaw angle θ is attached to the departure side is reduced by reducing the control gain (control amount) when the yaw angle θ is attached to the departure avoidance side. Vibration (hunting) is unlikely to occur, and the effect of preventing departure can be further increased.

(16) A control amount corresponding to the curvature of the travel lane L is added as the third term of the target control angle φL_Rt, φL_Lt by the travel lane edge control, that is, the control amount of the feedback control by the lateral displacement X.
As a result, even if the travel lane L is a curved road, the convergence with respect to the lateral displacement X with respect to the lateral displacement reference position is improved.

(17) A curve road correcting means is provided.
When the curvature ρ of the travel lane L is equal to or greater than a predetermined value, that is, a curved road, correction is performed with different control gains on the curve inner side and the curve outer side with respect to the center in the width direction of the travel lane L. That is, when the host vehicle C is located inside the curve of the travel lane L with respect to the center in the width direction of the travel lane L, the control gain is corrected to be smaller when the curvature is larger than when the curvature is small. On the other hand, when the host vehicle C is located outside the curve of the travel lane L with respect to the center in the width direction of the travel lane L, the control gain is corrected to be larger when the curvature is larger than when the curvature is small.
As a result, excessive control inside the curve can be prevented. That is, it is possible to reduce a sense of incongruity that occurs when the control amount is large, such as rebounding outside the curve.
In addition, insufficient control outside the curve can be prevented. That is, the control intervention when the yaw angle θ is generated to the outside of the curve is strengthened, and the effect of preventing departure is increased.

(18) First control amount calculation means for calculating a first control amount for reducing the lateral displacement deviation, and the traveling direction of the host vehicle with respect to the traveling lane when the host vehicle passes the lateral displacement threshold from the lane center side. And a third control amount calculating means for calculating a third control amount for reducing the angle deviation.
By this third control amount calculation means, the approach angle when the host vehicle enters the departure area outside the lateral displacement threshold can be reduced.
As a result of performing the yaw angle feedback control based on the third control amount, it is possible to keep the approach angle small when the host vehicle enters the departure area beyond the lateral displacement threshold from the traveling lane center side. Become. Thus, the yaw angle feedback control using the second control amount has an effect as a preliminary control for preventing departure.

In other words, when the approach angle is small, the deviation amount to the outside of the lateral displacement threshold value after the following own vehicle enters the deviation region becomes small. When the host vehicle enters the departure area, the first control amount used in the feedback control for reducing the lateral displacement deviation is reduced.
As a result, the feeling of being repelled from the traveling lane edge side is reduced, and the uncomfortable feeling as if the control is intervening on the departure side can be reduced. That is, it is possible to reduce the sense of restraint felt by the occupant.

(19) When the host vehicle is located in the departure side transition region on the lane center side of the lateral direction displacement threshold, the third control amount calculation means sets the control amount for reducing the angle deviation in the traveling direction of the host vehicle. calculate.
By performing yaw feedback with a control amount that reduces the angle deviation in the traveling direction of the host vehicle, the angle deviation (deviation in the traveling direction of the host vehicle progresses as it approaches the lateral displacement threshold from the center of the traveling lane. Direction) becomes smaller. As a result, it is possible to reduce the entry angle when entering the departure area from the center side exceeding the lateral displacement threshold.

(20) At this time, based on the control amounts of both the first control amount and the second control amount in at least a partial region (overlap control region) in the range on the lane edge side from the lateral displacement threshold. Control.
Thereby, if the own vehicle C is outside the lateral displacement threshold value on the left and right with respect to the traveling lane center Ls, feedback control of both the lateral displacement X and the yaw angle θ is performed. As a result, on the traveling lane end Le side, it is possible to perform the lane maintenance support in which the host vehicle C is appropriately kept in the traveling lane L and the feeling of being repelled from the traveling lane end Le is reduced.
That is, in the overlap control region on the lateral displacement threshold side in the departure region, both the control amount for the lateral position control and the control amount for the yaw angle control are used. As a synergistic effect using both control amounts, the following effects are obtained.

That is, when the vehicle advances with an angular deviation toward the departure side when entering the departure area, the direction of the control amount of the lateral position control and the direction of the control amount of the yaw angle control are the same direction (the departure avoidance direction). Control amount. As a result, the turning radius of the vehicle can be increased.
On the other hand, when the vehicle travels with an angular deviation toward the departure avoidance when heading for the lateral displacement threshold in the departure region, the direction of the control amount of the lateral position control and the control amount of the yaw angle control are opposite directions. It becomes the control amount. As a result, the trajectory of the vehicle gradually approaches the lateral displacement threshold, that is, the turning radius of the vehicle can be increased.
Thus, by adding the yaw angle control to the lateral position control, the turning radius of the vehicle for avoiding deviation can be increased. As a result, the acceleration / deceleration in the yaw direction can be reduced, and the rebound can be more reliably reduced.

(21) The control gain for the angle deviation is larger on the lane center side than on the lane end side.
Accordingly, the traveling direction of the vehicle is controlled in the direction along the traveling lane L on the traveling lane center Ls side while preventing an excessive control amount of the yaw angle control in the departure region.
(22) First control amount calculation means for calculating a first control amount for reducing the lateral displacement deviation of the host vehicle from the lateral displacement threshold, and the host vehicle from the wheel central portion side to the lateral displacement threshold. A fourth control amount calculating means is provided for calculating a fourth control amount for correcting the traveling locus in the approaching process so that the angle of the traveling direction of the host vehicle with respect to the traveling lane becomes smaller.

The fourth control amount calculating means corrects the travel locus in the process in which the host vehicle approaches the lateral displacement threshold from the wheel center side so that the angle of the traveling direction of the host vehicle with respect to the travel lane becomes smaller. As a result, the entry angle at which the host vehicle enters the departure area outside the lateral displacement threshold can be reduced.
When the approach angle is small, the amount of departure from the lateral displacement threshold to the outside after the following own vehicle enters the departure area becomes small. When the host vehicle enters the departure area, the first control amount used in the feedback control for reducing the lateral displacement deviation is reduced.
As a result, the feeling of being repelled from the traveling lane edge side is reduced, and the uncomfortable feeling as if the control is intervening on the departure side can be reduced. That is, it is possible to reduce the sense of restraint felt by the occupant.

(23) Fifth control for correcting the travel locus of the host vehicle according to the first control amount so as to reduce the angle of the traveling direction of the host vehicle with respect to the travel lane when the host vehicle approaches the lateral displacement threshold. Calculate the amount.
Thereby, in the departure area, when the own vehicle approaches the lateral displacement threshold in the departure avoidance direction, the turning radius of the own vehicle for departure avoidance can be increased. This leads to a reduction in the acceleration / deceleration in the yaw direction, and the feeling of being repelled can be reduced more reliably.

(24) The fourth control amount calculating means calculates a control amount for reducing the angle deviation of the traveling direction of the host vehicle with respect to the traveling lane when the host vehicle is located in the departure side transition region.
Accordingly, it is possible to calculate a control amount that makes the traveling locus in the process in which the host vehicle approaches the lateral displacement threshold from the wheel center side direction such that the angle of the traveling direction of the host vehicle with respect to the traveling lane becomes smaller.

(Modification)
(1) In the above-described embodiment, the case where the lateral displacement threshold value and the lateral displacement reference position match is illustrated. As shown in FIG. 22, the lateral displacement reference positions LXL and LXR may be set inside the lateral displacement threshold values LAL and LAR.
In this case, when the host vehicle C is located outside the lateral displacement threshold with respect to the travel lane center Ls, the vehicle C is directed toward lateral displacement reference positions LXL and LXR located inside the lateral displacement thresholds LAL and LAR. Feedback control is performed so that the lateral displacement X becomes smaller.
It becomes possible to adjust the control gain of the lateral feedback control.

(2) Further, when the lateral displacement reference positions LXL and LXR are set inside the lateral displacement threshold values LAL and LAR, the host vehicle C has the lateral displacement reference positions LXL and LXR and the lateral displacement threshold values LAL, Even in the case of being positioned between the LAR and the LAR, the feedback control may be performed so that the lateral displacement X becomes small. However, the control gain is kept small compared with the case where the host vehicle C is outside the lateral displacement thresholds LAL and LAR.
(3) In the above embodiment, the traveling lane edge Le is set as the lateral edge position. Alternatively, the lateral end position may be set inward by a predetermined amount from the traveling lane end Le. For example, it may be equal to the lateral displacement reference positions LXL and LXR.

(4) In the above embodiment, the weighting coefficients α_R, α_L, β_R, β_L have the relationship shown in FIG. 16, and relative to β_R, β_L with respect to α_R, α_L according to the lateral position of the host vehicle C. The size was changed.
The relationship among the weighting coefficients α_R, α_L, β_R, β_L is not limited to this.
For example,
α_R: β_R = 1: 1
α_L: β_L = 1: 1
And may be set constant. It has been confirmed that the effect can be obtained even with this setting.

(5) The position between the position offset from the lateral displacement threshold to the lane center side and the lateral displacement threshold is set as the departure side transition region. And in the area | region of the lane center part side from the departure transition area, the 2nd control amount is set to zero.
For example, as shown in FIG. 23, the weighting coefficients β_R and β_L are set to zero in a region on the lane center side with respect to the departure transition region.
As a result, the sense of restraint can be further reduced in the area closer to the center of the lane than the departure transition area.

(6) A position between the position offset from the lateral direction displacement threshold to the lane center side and the lateral direction displacement threshold is set as the departure side transition region. And in the area | region of the lane center part side from the departure transition area, the 3rd control amount is set to zero.
For example, as shown in FIG. 23, the weighting coefficients β_R and β_L are set to zero in a region on the lane center side with respect to the departure transition region.
As a result, the sense of restraint can be further reduced in the area closer to the center of the lane than the departure transition area.

(7) In the departure transition region, the control gain for the angle deviation is set larger on the lane center side than on the lane end side. Further, the control gain for the lateral displacement deviation is made smaller at the lane center side than at the lane end side.
For example, as shown in FIG. 23, the weighting coefficients β_R and β_L are set larger in the deviation transition region as they approach the lateral displacement thresholds XRt and XLt.
As a result, the effect of reducing the angular deviation in the departure direction is generated as the host vehicle C approaches the lateral displacement threshold values XRt and XLt. As a result, it is possible to reduce the entry angle when entering the departure area from the center side exceeding the lateral displacement threshold.

(8) In the above embodiment, the case where the traveling direction of the host vehicle is controlled by correcting the turning angle or the turning torque of the wheel based on the control amount is exemplified. Instead of correcting the turning angle or turning torque, the braking / driving amount or braking / driving force may be corrected based on the control amount. In this case, each control amount is calculated by the amount of yaw moment for reducing the lateral displacement deviation or the angle deviation. Then, each braking / driving force is corrected so as to generate a yaw moment corresponding to the control amount.
(9) The lateral displacement threshold may be only one in the left-right width direction. Alternatively, only one of the left and right lateral displacement threshold values may be set as the traveling lane edge position.
(10) In the above embodiment, the case where the front wheels are steered wheels has been illustrated. The rear wheels may be steered wheels that are steered according to the steering wheel.

(Second Embodiment)
Next, a second embodiment will be described with reference to the drawings. The same devices as those in the first embodiment will be described with the same reference numerals.
(Constitution)
The basic configuration of this embodiment is the same as that of the first embodiment. However, the process of step S200 in the lane keeping support controller 15 is different.
Other configurations are the same as those in the first embodiment.

The process of step S200 in the second embodiment will be described.
Step S200 is a processing unit that calculates final target turning angles φft and φrt. In the first embodiment, the left and right target turning angles φL_Lt and φL_Rt by the driving lane edge control calculated in step S180 and the left and right target turning angles by the driving lane center control calculated in step S190 are respectively weighted. In this state, the final target turning angles φft and φrt are calculated by summing.
On the other hand, in the present embodiment, φtL is selectively used by using the left and right target turning angle by the driving lane edge control calculated in step S180 and the left and right target turning angle by the driving lane center control calculated in step S190. , ΦtR are calculated, and the final target turning angles φft, φrt are calculated by taking the sum thereof.

Next, an example of the processing will be described.
First, the control amounts for the left and right are selectively calculated.
That is, the right control amount φL_Rt and φY_Rt are compared, and the larger value is defined as φtR. That is, φL_Rt and φY_Rt are taken as a select high to be φtR.
When φL_Rt is larger, Kfl and Krl are set to Kf_1 and Kr_1. On the other hand, when φY_Rt is larger, Kfy and Kry are set to Kf_1 and Kr_1.

Similarly, the left control amount φL_Lt and φY_Lt are compared, and the larger value is defined as φtL. That is, the select high of φL_Lt and φY_Lt is taken as φtL.
When φL_Lt is larger, Kfl and Krl are set to Kf_2 and Kr_2. On the other hand, when φY_Lt is larger, Kfy and Kry are set to Kf_2 and Kr_2.
Then, the final target turning angles φft and φrt are calculated by the following formula.
φft = Kf_1 × φtR + Kf_2 × φtL
φrt = Kr_1 × φtR + Kr_2 × φtL

(Action and operation)
When the final target turning angles φft, φrt are calculated, simply adding the lateral position feedback control amount and the yaw angle feedback control amount, the end side of the lane L when the yaw angle θ is large, etc. There is a problem that the control amount may be excessive. In order to solve this problem, a method of uniformly reducing the control gain can be considered. However, if the control gain of the yaw angle feedback is lowered, the control performance at the lane center Ls is lowered, and if the control gain of the lateral position feedback is lowered, the control performance at the lane edge Le is lowered.

For this reason, in the first embodiment, weighting is performed on the lateral position feedback control amount and the yaw angle feedback control amount, and the weighting is changed depending on the lateral position.
In the present embodiment, instead of this, select high is used.
That is, the selection amount of the right control amount (target turning angle φL_Rt, target turning angle φY_Rt) is performed. Similarly, select high of the left control amount (target turning angle φL_Lt, target turning angle φY_Lt) is performed. Thereafter, the final target turning angles φft and φrt are calculated by adding the left and right control amounts.

  As a result, when the host vehicle C is positioned within the left and right lateral displacement thresholds LXL and LXR, the final target turning angles φft and φrt are usually set to the target values based on the driving lane center control depending on the direction of the yaw angle θ. It becomes one value of the turning angle φY_Rt or the target turning angle φY_Lt. As a result, when the host vehicle C is located on the traveling lane center Ls side, yaw angle feedback control is performed.

In the following description, a case where the host vehicle C is located outside the right lateral displacement threshold values LXL and LXR will be described as an example.
First, it is assumed that the traveling direction of the host vehicle C is facing the departure side (right side).
At this time, the control amount on the right side is as follows.
That is, the target turning angle φL_Rt based on the traveling lane edge side control, which is the lateral position feedback control amount, is compared with the target turning angle φY_Rt based on the traveling lane center side control, which is the yaw angle feedback control amount. Then, based on the yaw angle θ and the lateral displacement X amount, if the target turning angle φL_Rt by the driving lane edge side control is larger, φtR becomes the target turning angle φL_Rt. On the other hand, if the target turning angle φY_Rt by the traveling lane center control is larger, φtR becomes the target turning angle φY_Rt.

On the other hand, the control amount on the left side is as follows.
Since ΔθL becomes zero, the target turning angle φY_Rt = 0 by the driving lane center portion control is obtained. Since ΔXL is also zero, the first term is zero based on equation (8). For this reason, φtL becomes the target turning angle φY_Lt by the traveling lane center side control, but is a small value.
For this reason, the final target turning angles φft, φrt are controlled by the larger value φtR of the right control amount.
Thereby, it is possible to prevent the control amount on the traveling lane edge Le side from becoming excessive.

Next, it is assumed that the direction of the traveling direction of the host vehicle C is toward the departure avoidance side (left side).
At this time, the control amount on the right side is as follows.
In other words, the target turning angle φY_Rt by the lane center Ls side control that is the yaw angle feedback control amount becomes zero. For this reason, the target turning angle φL_Rt by the lane edge Le side control, which is the lateral position feedback control amount, becomes φtR.

On the other hand, the control amount on the left side is as follows.
Since ΔXL is also zero, the first term is zero based on equation (8). For this reason, φtL becomes the target turning angle φY_Lt by the lane center Ls side control, but is a small value.
For this reason, if ΔθL, which is the yaw angle θ to the left side, is equal to or greater than a predetermined amount, the target turning angle φY_Rt by the traveling lane center portion control becomes φtL. If ΔθL, which is the yaw angle θ to the left, is small, φtL is a small value.
For this reason, the final target turning angles φft and φrt can be repelled from the right travel lane edge Le and the feeling of reduction can be reduced.

(Effect of this embodiment)
(1) The larger one of the first control amount and the second control amount (third control amount, fourth control amount) is set as the final control amount.
That is, after selecting the lateral position feedback control amount and the yaw angle feedback control amount separately for the left and right, the final target turning angles φft and φrt are calculated by adding the left and right control amounts.
As a result, it is possible to ensure the control performance on the travel lane edge Le side while ensuring the control performance on the travel lane center Ls side without excessively increasing the control amounts of the final target turning angles φft and φrt.
(2) Other effects are the same as those of the first embodiment.

(Modification)
(1) The following threshold processing may be added to the above selection and addition processing in which select high is performed separately on the left and right.
That is, when taking the select amount of the control amount on the right side,
If φL_Rt> φY_Rt, as described above,
Let φtR = φL_Rt.

At this time, if φY_Rt> φth_Y, that is, if the target turning angle φY_Rt by the traveling lane center portion control is equal to or larger than a predetermined threshold, the correction is performed as shown in the following equation.
φtR = φtR + K1 × (φY_Rt−φth_Y)
Conversely, if φL_Rt ≦ φY_Rt, as described above,
Let φtR = φY_Rt.

At this time, if φL_Rt> φth_L, that is, if the target turning angle φL_Rt by the traveling lane edge control is equal to or larger than a predetermined threshold value, the correction is performed according to the following equation.
φtR = φtR + K2 × (φL_Rt−φth_L).
Similarly, when taking the select amount of the control amount on the left side,
If φL_Lt> φY_Lt, as described above,
Let φtL = φL_Lt.

At this time, if φY_Lt <−φth_Y, that is, if the target turning angle φY_Lt by the traveling lane center portion control is greater than or equal to a predetermined threshold value, correction is performed as in the following equation.
φtL = φtL + K3 × (φY_Lt + φth_Y).
Conversely, if φL_Lt ≦ φY_Lt, as described above,
Let φtL = β_L × φY_Lt.

At this time, if φL_Lt <−φth_L, that is, if the target turning angle φL_Lt by the traveling lane edge control is greater than or equal to a predetermined threshold value, the correction is performed according to the following equation.
φtL = φtL + K4 × (φL_Lt + φth_L).
Then, final target turning angles φft and φrt are calculated by the following formula.
φft = Kf_1 × φtR + Kf_2 × φtL
φrt = Kr_1 × φtR + Kr_2 × φtL
As described above, when the left and right select low sides are larger than the predetermined amount, the corresponding correction is performed.
In this way, control performance is ensured even when the lateral displacement X and yaw angle θ are large on the lane edge Le side without excessively increasing the control amounts of the final target turning angles φft and φrt. it can.

(2) Also, after multiplying several types of operations by weighting factors, select the one with the largest control amount, and set the control amount in the left direction and the control amount in the right direction. It may be a control amount. In this way, it is possible to perform control more suitable for the driving scene by adjusting the ease of selecting each control amount according to the driving state, road information, and the like.

An example of the processing is shown below.
When making a comparison and selection, gains γ_yaw and γ_lat determined from information about the traveling vehicle C such as the yaw angle θ, lateral position, curvature, navigation information, and road information are used to weight the selection amount.
Here, γ_yaw is a gain that becomes large in a scene with a large deviation risk with respect to the yaw angle θ, such as when a yaw angle θ deviating to the outside of the curve has occurred. γ_lat is a gain that increases in a scene where there is a large deviation risk relative to traveling in the current lateral position, such as when traveling outside the curve at the yaw angle θ.

And
If γ_lat × φL_Rt> γ_yaw × φY_Rt,
If φtR = φL_Rt, while γ_lat × φL_Rt ≦ γ_yaw × φY_Rt,
Let φtR = φY_Rt.
Similarly, if γ_lat × φL_Lt> γ_yaw × φY_Lt,
φtL = φL_Lt On the other hand, if γ_lat × φL_Lt ≦ γ_yaw × φY_Lt,
Let φtL = φY_Lt.
Then, final target turning angles φft and φrt are calculated by the following formula.
φft = Kf_1 × φtR + Kf_2 × φtL
φrt = Kr_1 × φtR + Kr_2 × φtL

(3) Also, the curve is determined from the curvature of the lane L, the control amount from the outside of the curve to the inside is the addition of the respective calculation amount, and the control amount from the inside of the curve to the outside is selected with the largest control amount, You may make it add them and make it a lane maintenance control amount.
In this way, the control amounts for the deviation to the outside of the curve are added to the respective control amounts, and the control amount for the deviation to the inside of the curve is selected high. As a result, it is possible to increase only the control amount for the deviation to the outside of the curve having a high risk (requires a large control amount). In other scenes, the control performance at the end of the travel lane L can be ensured while ensuring the control performance at the center Ls of the travel lane without excessively increasing the control amounts of the final target turning angles φft and φrt.

An example of the processing is shown below.
Using the road curvature ρ, the control amount is added to the deviation to the outside of the curve as described below, and the deviation to the inside of the curve is selectively processed.
When ρ <0 (right curve), calculation is performed as follows.
That is, for the right control amount, if φL_Rt> φY_Rt,
Let φtR = φL_Rt.
On the other hand, if φL_Rt ≦ φY_Rt, φtR = φY_Rt.

For the left control amount, an added value is taken as in the following equation.
φtL = φL_Lt + φY_Lt
Conversely, when ρ> 0 (left curve), the calculation is performed as follows.
That is, for the control amount on the right, an added value is taken as in the following equation.
φtR = φL_Rt + φY_Rt
For the left control amount, if φL_Lt> φY_Lt,
φtL = α_L × φL_Lt.
On the other hand, if φL_Lt ≦ φY_Lt,
Let φtL = β_L × φY_Lt.

Further, when ρ = 0, select high is performed separately for the left and right as follows.
For the control amount on the right side,
If φL_Rt> φY_Rt, then φtR = φL_Rt.
If φL_Rt ≦ φY_Rt, then φtR = φY_Rt.
For the control amount on the left side,
If φL_Lt> φY_Lt, then φtL = φL_Lt.
If φL_Lt ≦ φY_Lt, then φtL = φY_Lt.
Then, final target turning angles φft and φrt are calculated by the following formula.
φft = Kf_1 × φtR + Kf_2 × φtL
φrt = Kr_1 × φtR + Kr_2 × φtL

5 Steering Actuator 11 Steering Controller 12 Steering Wheel 13 Front Wheel 14 Own Vehicle State Parameter 15 Lane Keeping Support Controller 16 External Recognizing Means 17 Direction Indicator Switch 25 Rear Wheel Steering Actuator C Own Vehicle Kc_L1 Correction Gain Kc_L2 Correction Gain Kdl Correction Gain Kdy correction gain Kel correction gain Key correction gain Kfl gain Kfy front wheel steering gain Kfy front wheel steering gain Kry rear wheel steering gain Ksl correction gain Ksy correction gain Kvl correction gain Kvy correction gain Kvy correction gain Ky_L feedback gain Ky_R feedback gain Kρ_L Lane end control feedback correction gain KρL_R Travel lane end control feedback correction gain KρY_L Travel lane center control feed Correction correction gain KρY_R Travel lane center portion control feedback correction gain L Travel lane Le Left travel lane edge Ls Travel lane center LXL Lateral displacement reference position Vw Wheel speed Wlane Travel lane width X Lateral displacement XLt Lateral displacement reference threshold XRt Lateral displacement Reference threshold α_R coefficient β_R coefficient γ_yaw gain δ steering angle ΔXL lateral displacement deviation ΔXR lateral displacement deviation ΔθL yaw angle deviation ΔθR yaw angle deviation θ yaw angle ρ curvature φft final target turning angle for front wheels φrt final target turning angle for rear wheels

Claims (15)

Lateral displacement acquisition means for acquiring information on the lateral displacement of the host vehicle relative to the traveling lane;
A lateral displacement threshold value setting means for providing a lateral displacement threshold value in a travel lane in which the host vehicle travels
A first control amount calculating means for calculating a first control amount for reducing a lateral displacement deviation of the host vehicle from the lateral displacement threshold; and a second control unit for reducing an angle deviation of the traveling direction of the host vehicle with respect to the travel lane. A second control amount calculating means for calculating a control amount;
When the host vehicle passes the lateral displacement threshold value from the width direction center side of the travel lane, the vehicle travels more than the second control amount and the lateral displacement threshold value on the travel lane center side from the lateral displacement threshold value. Travel direction control means for controlling the turning angle or turning torque of the front and rear wheels based on at least the first control amount on the outer side in the width direction of the lane;
With
The advancing direction control means controls the phase of the steering direction of the rear wheels relative to the steering direction of the front wheels in accordance with at least one of the first control amount and the second control amount. have a rudder direction adjustment means,
The steered direction adjusting means steers the rear wheels in a direction opposite to the steered direction of the front wheels according to the second control amount, and turns the rear wheels with respect to the second control amount. A lane keeping assist device that reduces the amount of steering of the front wheels in accordance with an increase in rudder . Lateral displacement acquisition means for acquiring information on the lateral displacement of the host vehicle relative to the traveling lane;
A lateral displacement threshold value setting means for providing a lateral displacement threshold value in a travel lane in which the host vehicle travels;
A first control amount calculating means for calculating a first control amount for reducing a lateral displacement deviation of the host vehicle from the lateral displacement threshold; and a second control unit for reducing an angle deviation of the traveling direction of the host vehicle with respect to the travel lane. A second control amount calculating means for calculating a control amount;
When the host vehicle passes the lateral displacement threshold value from the width direction center side of the travel lane, the vehicle travels more than the second control amount and the lateral displacement threshold value on the travel lane center side from the lateral displacement threshold value. Travel direction control means for controlling the turning angle or turning torque of the front and rear wheels based on at least the first control amount on the outer side in the width direction of the lane;
With
The advancing direction control means controls the phase of the steering direction of the rear wheels relative to the steering direction of the front wheels in accordance with at least one of the first control amount and the second control amount. have a rudder direction adjustment means,
The traveling direction control means calculates a final control amount from a sum of a value obtained by multiplying the first control amount by a first weighting factor and a value obtained by multiplying the second control amount by a second weighting factor. ,
The first weighting coefficient and the second weighting coefficient are changed according to the lateral displacement deviation, and the larger the lateral displacement deviation, the larger the first weighting coefficient is set with respect to the second weighting coefficient. . Lateral displacement acquisition means for acquiring information on the lateral displacement of the host vehicle relative to the traveling lane;
A lateral displacement threshold value setting means for providing a lateral displacement threshold value in a travel lane in which the host vehicle travels;
A first control amount calculating means for calculating a first control amount for reducing a lateral displacement deviation of the host vehicle from the lateral displacement threshold; and a second control unit for reducing an angle deviation of the traveling direction of the host vehicle with respect to the travel lane. A second control amount calculating means for calculating a control amount;
When the host vehicle passes the lateral displacement threshold value from the width direction center side of the travel lane, the vehicle travels more than the second control amount and the lateral displacement threshold value on the travel lane center side from the lateral displacement threshold value. Travel direction control means for controlling the turning angle or turning torque of the front and rear wheels based on at least the first control amount on the outer side in the width direction of the lane;
With
The advancing direction control means controls the phase of the steering direction of the rear wheels relative to the steering direction of the front wheels in accordance with at least one of the first control amount and the second control amount. have a rudder direction adjustment means,
The lane keeping assist device according to claim 1, wherein the traveling direction control means sets the larger one of the first control amount and the second control amount as a final control amount . The steering direction adjustment means, in response to the first control amount, to any one of claims 1 to 3, characterized in that steers the rear wheels to the front wheels in the steering direction and phase direction Lane maintenance support device described in 1. The lane keeping assist device according to claim 4 , wherein the turning amount of the front wheel is increased with respect to the first control amount in accordance with an increase of the turning amount of the rear wheel.   The lane keeping assist device according to any one of claims 1 to 5, wherein the turning direction adjusting means corrects the turning of the rear wheel in accordance with a traveling environment around the vehicle.   The lane keeping assist device according to any one of claims 1 to 6, wherein the turning direction adjusting means corrects the turning of the rear wheel according to a vehicle speed.   The steering direction adjusting means corrects the steering of the rear wheel in accordance with a driver steering intention estimation value, and suppresses the steering of the rear wheel when it is estimated that there is a steering intention. The lane keeping assist device according to any one of claims 7 to 9.   The advancing direction control means performs control based on the control amounts of both the first control amount and the second control amount in at least a partial region in the range on the lane edge side from the lateral direction displacement threshold. The lane keeping assist device according to any one of claims 1 to 8, wherein the lane keeping assist device is characterized in that Set the lateral displacement reference position within two lateral displacement thresholds that are individually offset from the center in the width direction of the driving lane to the left and right in the width direction,
The first control amount calculating means calculates a first control amount using a lateral displacement deviation from a lateral displacement reference position on a side closer to the own vehicle among the left and right lateral displacement reference positions, 10. The method according to claim 1, wherein when it is determined that the position is between the left and right lateral displacement thresholds, the lateral displacement deviation is regarded as zero, or a control gain for the lateral displacement deviation is reduced. The lane keeping assist device according to any one of the above. The lane keeping assist device according to any one of claims 1 to 10 , wherein the first control amount is corrected by a lateral displacement speed of the host vehicle. The control gain of the second control amount is corrected by the distance of the host vehicle with respect to the traveling lane end located on the traveling direction side of the own vehicle among the left and right traveling lane ends, and the own vehicle relative to the traveling lane end is corrected. The lane keeping assist device according to any one of claims 1 to 11 , wherein the control gain is corrected such that the control gain increases as the distance is shorter. When the curvature of the traveling lane is equal to or greater than a predetermined value, the vehicle includes a curve road correcting unit that corrects a control gain for the control amount calculated by the control amount calculating unit based on the position of the host vehicle with respect to the center of the traveling lane and the curvature of the traveling lane,
The correction of the control gain is performed by correcting the control gain to be smaller when the curvature is larger than when the curvature is small when the host vehicle is positioned inside the curve of the traveling lane with respect to the center in the width direction of the traveling lane. The control gain is corrected to be larger when the curvature is larger than when the curvature is small when the host vehicle is located outside the curve of the traveling lane with respect to the center in the width direction of the lane. Item 13. A lane keeping assist device according to any one of items 12 . The distance between the lateral displacement threshold and the position offset to the lane center side and the lateral displacement threshold is the departure side transition region,
The lane keeping assist device according to any one of claims 1 to 13 , wherein the second control amount is set to zero in a region closer to the center of the lane than the departure transition region. While providing a lateral displacement threshold for the travel lane on which the host vehicle travels, obtain information on the lateral displacement of the host vehicle relative to the travel lane,
When the host vehicle passes the lateral displacement threshold value from the center side in the width direction of the travel lane, the angle deviation in the traveling direction of the host vehicle with respect to the travel lane is smaller on the travel lane center side than the lateral displacement threshold value. The traveling direction of the host vehicle is controlled such that the lateral displacement deviation of the host vehicle from at least the lateral displacement threshold is smaller at the outer side in the width direction of the travel lane than the lateral displacement threshold. Control
In the control to reduce the angle deviation, the steering direction of the rear wheels is set to the opposite phase to the steering direction of the front wheels, and the amount of steering of the front wheels is reduced according to the increase in the steering of the rear wheels. Lane maintenance support method characterized by this.

Images