JP2019059449A - Vehicle driving support device - Google Patents

Abstract

To provide a vehicle driving support device capable of lowering a possibility that a determination on whether a steering tracking target driver's vehicle deviates from a travel lane decreases in precision.SOLUTION: A driving support ECU comprises a Kalman filter 10a. The driving support ECU is configured to input to the Kalman filter (10a) a first distance (DL1) and a first yaw angle (θ1) as observation values, a first standard deviation (σ1) as process noise, and a second standard deviation (σ2) and a third standard deviation (σ3) as observation noise, and to output from the Kalman filter (10a) a first yar angle variation rate (θ1'), an estimated first distance (DL1f), and an estimated first yaw angle (θ1f) as output values.SELECTED DRAWING: Figure 6

Description

  The present invention relates to a vehicle driving support device that performs steering control for causing a host vehicle to travel along a target travel line based on the travel locus of another vehicle (front vehicle) traveling in the front region of the host vehicle.

  One of the conventionally known vehicle driving support devices (hereinafter referred to as "conventional devices") generates a traveling trace of a steering following target vehicle selected from among preceding vehicles traveling in front of the own vehicle. . The conventional device performs steering control (steering follow control) so that the host vehicle travels along a target travel line based on the generated travel locus.

  When steering following control is performed, if the target vehicle following the steering deviates from the traveling lane (the traveling lane), there is a high possibility that the traveling locus is largely deviated from the vicinity of the center of the traveling line. Therefore, in this case, it is not preferable for the conventional device to set the target travel line using the generated travel locus. For this reason, the conventional device determines whether or not the steering following target vehicle deviates from the traveling lane as follows, and when the steering following target vehicle deviates from the traveling lane: Stop steering following control.

  In the conventional device, a steering following target vehicle at the time when the steering following target vehicle exists at a position on the traveling locus corresponding to the current position of the own vehicle (hereinafter referred to as "own vehicle corresponding position"). The positional relationship between and and the white line is acquired using the traveling trace of the target vehicle following the steering.

  That is, in the conventional device, first, the positional relationship between the steering target vehicle in the past when it was present at the position corresponding to the vehicle (hereinafter simply referred to as "the target vehicle for tracking the past steering") and the white line. A first distance DL1 in the lane width direction between the vehicle corresponding position and the white line, and a first yaw angle θ1 with respect to the white line of the target vehicle following the steering in the past are acquired. The first yaw angle θ1 is calculated by subtracting the fourth yaw angle θ4 with respect to the white line of the host vehicle from the third yaw angle θ3 with respect to the traveling locus of the target vehicle following the steering of the host vehicle.

Next, according to the following formula, the conventional apparatus estimates the arrival time Tx1 until the target vehicle following the steering reaches the white line and the movement required for the target vehicle following the steering to move from the position corresponding to the vehicle to the current position The time Tx2 is calculated.
Expected arrival time Tx1 = “first distance DL1 / estimated lateral speed (steering follow target vehicle speed v × sin θ1)”
Travel time Tx2 = “inter-vehicle distance / steering follow target vehicle speed v”

  Further, the conventional apparatus calculates the determination time Tx (= the arrival expected time Tx1−the movement time Tx2) by subtracting the moving time Tx2 from the determination time Tx = the expected arrival time Tx1. When the determination time Tx is smaller than the predetermined threshold value, the conventional device determines that the steering follow-up target vehicle is in a situation of departing from the traveling lane, and stops the steering follow-up control (see, for example, Patent Document 1).

JP, 2016-101783, A

  The conventional apparatus is premised on the fact that the yaw angle to the white line of the steering following target vehicle is constant and does not change from the first yaw angle θ1 while the steering following target vehicle moves from the vehicle corresponding position to the current position. And the lane departure of the target vehicle following the steering is determined.

  However, in practice, it is possible that the yaw angle with respect to the white line of the steering following target vehicle may be changed from the first yaw angle θ1 while the steering following target vehicle is moving from the vehicle corresponding position to the current position. . In this case, since the accuracy of the determination time Tx is lowered due to the decrease in the accuracy of the estimated arrival time Tx1, the accuracy of the determination of the travel lane deviation of the steering following target vehicle is lowered.

On the other hand, the inventor of the present application examined the following vehicle driving support device.
That is, this vehicle driving support device is configured to change the first yaw angle θ1 per unit time of the steering following target vehicle with respect to the white line of the steering following target vehicle θ1 ′ (“first yaw angle change rate” or “yaw angular velocity” Is called)). Then, using the first yaw angle change rate θ ′, it is determined whether or not the steering following target vehicle deviates from the traveling lane.

  However, when the first yaw angle change rate θ ′ is obtained by calculating from the change of the actual first yaw angle θ1, the obtained first yaw angle change rate θ ′ is very noisy and the accuracy is low. I will. Therefore, there is a possibility that the determination accuracy as to whether or not the steering following target person vehicle deviates from the traveling lane.

  The present invention was made to address the problems described above. That is, one of the objects of the present invention is to provide a vehicle driving support device (hereinafter referred to as “the present invention,” which can reduce the possibility that the determination accuracy of whether or not the steering following target person vehicle deviates from the traveling lane. Also referred to as “device”.

The apparatus according to the present invention comprises a lane marking recognition unit (17b) for recognizing lane markings of a traveling lane in which a host vehicle (SV) is traveling;
A marking line recognition accuracy judgment unit (10) which judges the recognition accuracy of the marking line;
A lane width acquisition unit (10) for acquiring the lane width (W) of the traveling lane based on the recognized lane line;
A traveling locus generating unit (10) for generating a traveling locus (L1) of a steering following target vehicle (TV) specified from among preceding vehicles traveling ahead of the host vehicle;
A vehicle speed acquisition unit (17a, 10) for acquiring the vehicle speed (v) of the target vehicle following the steering;
Each time the predetermined time (Δtcy) elapses, the target vehicle follows in the past when the longitudinal distance is the same as the host vehicle and the lateral position exists at the host vehicle corresponding position on the traveling track. A first distance (DL1) in a lane width direction between the marking line (LW) of (TVp) and the past steering follow-up target vehicle, and a first yaw to the marking line of the past steering follow-up target vehicle The angle (θ1),
A third distance (DL3) in the lane width direction between the host vehicle and the host vehicle corresponding position, acquired based on the recognized dividing line and the travel locus, and the steering follow of the host vehicle The third yaw angle (θ3) of the target vehicle with respect to the traveling locus, the fourth distance (DL4) in the lane width direction between the vehicle and the lane line, and the fourth distance to the lane line of the vehicle A parameter acquisition unit to acquire based on the yaw angle (θ4);
The first standard deviation (σ1) is calculated by substituting the vehicle speed of the target vehicle following the steering and the lane width into the following calculation formula 1, and the second standard deviation (σ2) is calculated using the following calculation formula 1 The standard deviation of the error of the distance is calculated (step 945), and the standard deviation of the error of the first yaw angle which is the third standard deviation (σ3) is calculated using the following formula 3 (step 950) (10) and
When it is determined by the marking recognition accuracy determination unit that the marking recognition accuracy is high, the standard deviation of the error of the fourth distance in the calculation formula 2 is set to a first value, and in the calculation formula 3, Set the standard deviation of the 4th yaw angle error to 3
When it is determined by the marking recognition accuracy determination unit that the marking recognition accuracy is low, the standard deviation of the error of the fourth distance in the calculation formula 2 is set to the second value larger than the first value. The standard deviation calculating unit configured to set the standard deviation of the error of the fourth yaw angle in the formula 3 to a fourth value larger than the third value (step 950);
A parameter estimation unit (10) provided with a Kalman filter (10a),
The obtained first distance and the first yaw angle are input to the Kalman filter as observation values each time the predetermined time passes.
Input the first standard deviation as process noise,
Input the second standard deviation and the third standard deviation as observation noise,
According to the principle of the Kalman filter, a first yaw angle change rate (θ1 ′) which is a change amount of the first yaw angle per predetermined time, an estimated first distance (DL1f), and an estimated first yaw angle (θ1f) The parameter estimation unit configured to calculate the first yaw angle change rate, the estimated first distance, and the estimated first yaw angle calculated as the output value from the Kalman filter;
A first parameter including the first yaw angle change rate, the estimated first distance, and the estimated first yaw angle, and a second with respect to the division line of the steering following target vehicle calculated using the first parameter A second distance (DL2) in the lane width direction between the steering following target vehicle and the lane line using at least one of the parameters including the yaw angle (θ2) and the vehicle speed of the steering following target vehicle An estimated lateral speed (v2) of the steering following target vehicle, and a predicted time (Ta) until the steering following target vehicle reaches the lane line, and the calculated second distance, estimated lateral speed, And based on at least one of the prediction times,
Determining whether the steering following target vehicle is in a departure situation deviating from the travel lane (step 965); a travel lane departure situation determination unit (10);
A travel control unit (10) that executes a steering follow-up control that changes a steering angle of the host vehicle so that the host vehicle travels along a target travel line set based on the travel locus;
Equipped with
The travel control unit stops the steering following control (step 910) when it is determined that the steering following target vehicle is in the deviation situation (determination of “Yes” in step 965).
If it is determined that the steering follow-up target vehicle is not in the departure situation (a determination of “No” in step 965), the steering follow-up control is executed (step 970).
Configured as.
(Calculation formula 1)
1st standard deviation = (4Vxmax ³ / W ² ) x (Δtcy / 3v)
(First standard deviation: Standard deviation of the change amount of the first yaw angle change rate per predetermined time, Vxmax: maximum lateral velocity, W: the lane width, Δty: the predetermined time, v: the steering following target vehicle Speed of
(Calculation formula 2)
(Standard deviation of the first distance error) = ((standard deviation of the third distance error) ² + (standard deviation of the fourth distance error) ² ) ^0.5
(Calculation formula 3)
(Standard deviation of error of first yaw angle) = ((standard deviation of error of third yaw angle) ² + (standard deviation of error of fourth yaw angle) ² ) ^0.5

  According to this, the value of the first standard deviation input as the process noise to the Kalman filter is determined in accordance with the vehicle speed of the target vehicle following the steering and the lane width of the traveling lane. Furthermore, in accordance with the recognition accuracy of the white line, the values of the second standard deviation and the third standard deviation input as observation noise are determined. As a result, the first yaw angle change rate used for calculating the determination parameter can be accurately obtained, and the accuracy of the determination parameter can be improved. As a result, it is possible to reduce the possibility that the determination accuracy as to whether or not the steering following target person vehicle deviates from the traveling lane is lowered.

  In the above description, in order to facilitate understanding of the present invention, the names and / or symbols used in the embodiments are attached in parentheses to the configuration of the invention corresponding to the embodiments described later. However, each component of the present invention is not limited to the embodiment defined by the above-mentioned name and / or code.

FIG. 1 is a schematic block diagram of a vehicle driving support device according to an embodiment of the present invention. FIG. 2 is a plan view for explaining the lane keeping control. FIG. 3A is a plan view for explaining the lane keeping control. FIG. 3B is an equation for explaining the relationship between the coefficient of the cubic function of the traveling locus and the curvature and the like. FIG. 3C is an equation for explaining the relationship between the coefficient of the cubic function of the traveling locus and the curvature and the like. FIG. 4A is a plan view of a road and a vehicle for illustrating the operation of the vehicle driving support device according to the embodiment of the present invention. FIG. 4 (B) is a plan view of a road and a vehicle for explaining the operation of the vehicle driving support device according to the embodiment of the present invention. FIG. 5 is a plan view of a road and a vehicle for explaining the operation of the vehicle driving support device according to the embodiment of the present invention. FIG. 6 is a schematic view showing the operation of the Kalman filter. FIG. 7 is a graph for explaining the derivation method of (calculation formula 1). FIG. 8 is a plan view of a road and a vehicle for illustrating the operation of the vehicle driving support device according to the embodiment of the present invention. FIG. 9 is a flowchart showing a routine executed by the CPU of the driving support ECU provided in the vehicle driving support device according to the embodiment of the present invention.

  Hereinafter, a vehicle driving support device (hereinafter, also referred to as a “present implementation device”) according to an embodiment of the present invention will be described with reference to the drawings. The present implementation device is also a vehicle travel control device. In all the drawings of the embodiment, the same or corresponding parts are denoted by the same reference numerals.

(Constitution)
The present embodiment is applied to a vehicle (automobile) as shown in FIG. The vehicle to which the present implementation device is applied may be referred to as "own vehicle" in order to distinguish it from other vehicles. The present implementation device includes a driving support ECU 10, an engine ECU 30, a brake ECU 40, a steering ECU 60, a meter ECU 70, an alarm ECU 80, and a navigation ECU 90. In the following, the driving support ECU 10 is also simply referred to as “DSECU”.

  These ECUs are electric control units (Electric Control Units) each having a microcomputer as a main part, and are connected to be able to transmit and receive information mutually via a CAN (Controller Area Network) not shown. The microcomputer includes a CPU, a ROM, a RAM, a non-volatile memory, an interface I / F, and the like. The CPU implements various functions by executing instructions (programs, routines) stored in the ROM. Some or all of these ECUs may be integrated into one ECU.

  The DSECU is connected to the sensors (including switches) listed below to receive detection signals or output signals of those sensors. Each sensor may be connected to an ECU other than the DSECU. In that case, the DSECU receives the detection signal or output signal of the sensor from the ECU to which the sensor is connected via the CAN.

The accelerator pedal operation amount sensor 11 detects an operation amount (accelerator opening degree) of the accelerator pedal 11a of the host vehicle, and outputs a signal representing an accelerator pedal operation amount AP.
The brake pedal operation amount sensor 12 detects an operation amount of the brake pedal 12a of the host vehicle, and outputs a signal representing the brake pedal operation amount BP.

The steering angle sensor 14 detects a steering operation angle which is a rotation angle of the steering wheel SW of the host vehicle, and outputs a signal representing the steering operation angle θ.
The steering torque sensor 15 detects a steering torque applied to the steering shaft US of the host vehicle by the operation of the steering wheel SW, and outputs a signal representing the steering torque Tra.
The vehicle speed sensor 16 detects the traveling speed (vehicle speed) of the host vehicle and outputs a signal representing the vehicle speed SPD.

  The ambient sensor 17 includes a radar sensor 17a, a camera sensor 17b, and a target recognition unit 17c. The surrounding sensor 17 is configured to obtain information on at least a road ahead of the host vehicle and a three-dimensional object present on the road. The three-dimensional object represents, for example, moving objects such as pedestrians, bicycles and automobiles, and fixed objects such as telephone poles, trees and guardrails. Hereinafter, these three-dimensional objects may be referred to as "targets".

  The surrounding sensor 17 detects the presence or absence of a target based on the information detected from a three-dimensional object by at least one of the radar sensor 17a and the camera sensor 17b, the target ID of the recognized target (n), the vertical distance Dfx (n) , Information on a target (n) including the lateral position Dfy (n), the relative velocity Vfx (n), the relative lateral velocity Vfy (n), the vehicle speed v, etc. (hereinafter referred to as "target information") It is designed to calculate and output.

  In addition, the surrounding sensor 17 acquires these values based on the predetermined xy coordinate (refer FIG. 2). The x-axis extends along the front-rear direction of the host vehicle SV so as to pass through the widthwise center position of the front end of the host vehicle SV, and is a coordinate axis having the front as a positive value. The y-axis is a coordinate axis orthogonal to the x-axis and having the left direction of the host vehicle SV as a positive value. The origin of the x axis and the origin of the y axis are center positions in the width direction of the front end of the vehicle SV. The x-coordinate position of the x-y coordinate is called a vertical distance Dfx, and the y-coordinate position is called a horizontal position Dfy.

The vertical distance Dfx (n) of the target (n) is the front end of the host vehicle SV and the rear end of the target (n) (for example, a front vehicle which is another vehicle traveling in the front region of the host vehicle SV) The signed distance in the central axis direction (x-axis direction) of the host vehicle SV between them.
The lateral position Dfy (n) of the target (n) is orthogonal to the central axis of the vehicle SV at "the central position of the target (n) (for example, the center position in the vehicle width direction of the rear end of the preceding vehicle)". Is the signed distance in the direction (y-axis direction).
The relative velocity Vfx (n) of the target (n) is the difference (= Vs-Vj) between the velocity Vs of the target (n) and the velocity Vj (= SPD) of the host vehicle SV. The velocity Vs of the target (n) is the velocity of the target (n) in the central axis direction (x-axis direction) of the host vehicle SV.
The relative lateral velocity Vfy (n) of the target (n) is the velocity (signed velocity) in the direction (y-axis direction) orthogonal to the central axis of the vehicle SV at the central position of the target (n) .

  The radar sensor 17a shown in FIG. 1 includes a radar wave transmitting / receiving unit and a processing unit. The radar wave transmitting / receiving unit emits, for example, a millimeter wave band radio wave (hereinafter referred to as “millimeter wave”) at least in the peripheral region of the vehicle SV including at least the front region of the vehicle SV. A wave receives a reflected wave generated by being reflected by a portion (immediately, a reflection point) of a solid. The radar sensor 17a may be a radar sensor that uses radio waves (radar waves) in frequency bands other than the millimeter wave band.

  The processing unit of the radar sensor 17a uses reflection point information including the phase difference between the transmitted millimeter wave and the received reflected wave, the attenuation level of the reflected wave, the time from the transmission of the millimeter wave to the reception of the reflected wave, and the like. Based on the determination of the presence or absence of a target.

  Furthermore, the processing unit of the radar sensor 17a determines the vertical distance Dfx of the target, the heading θp of the target with respect to the host vehicle SV, and the host vehicle SV, based on the reflection point information of the reflection points belonging to the recognized target. The relative velocity Vfx with the target, the vehicle speed v of the target, and the like (hereinafter referred to as “radar sensor detection information”) are calculated.

  The camera sensor 17 b includes a stereo camera and an image processing unit. The stereo camera captures the scenery of the “left side area and the right side area” in front of the host vehicle SV and acquires a pair of left and right image data.

The image processing unit determines the presence or absence of a target present in the imaging area based on the photographed left and right pair of image data. When it is determined that a target exists, the image processing unit determines the azimuth θ _p of the target, the vertical distance Dfx of the target, the relative velocity Vfx between the vehicle SV and the target, etc. Camera sensor detection information) is calculated.

  The target recognition unit 17 c is connected in a communicable state with the processing unit of the radar sensor 17 a and the image processing unit of the camera sensor 17 b, and receives “radar sensor detection information” and “camera sensor detection information”. It has become. The target recognition unit 17c uses the “target ID, vertical distance Dfx (n) of the target (n) recognized using at least one of the“ radar sensor detection information ”and the“ camera sensor detection information ”, and the horizontal distance Target information including the position Dfy (n) and the relative velocity Vfx (n) is determined (acquired). The target recognition unit 17c transmits the determined target information to the DSECU each time a predetermined time has elapsed.

  Furthermore, the image processing unit of the camera sensor 17b is based on the pair of left and right image data, lane markings (such as lane markers, hereinafter simply referred to as "white lines") such as white lines on the left and right of the road. recognize. Then, the image processing unit determines the shape (for example, curvature radius) of the host vehicle SV traveling lane, which is the lane in which the host vehicle SV is traveling, and the positional relationship between the host vehicle SV traveling lane and the host vehicle SV for a predetermined time. It is calculated and sent to DSECU every time e. The positional relationship between the host vehicle SV traveling lane and the host vehicle SV is, for example, a lane between a central position (that is, a central line) of the left white line and the right white line of the host vehicle traveling lane and the central position in the vehicle width direction of the host vehicle SV. It is represented by the distance in the width direction, the angle between the direction of the center line and the x-axis direction of the vehicle SV (that is, the yaw angle), or the like.

  Furthermore, the camera sensor 17b calculates information (in some cases, referred to as "recognition distance information") indicating how many meters ahead of the front white line the camera sensor 17b recognizes and outputs it to the DSECU. It has become. The DSECU is configured to determine whether the white line recognition accuracy is high based on the distance (recognition distance information) at which the camera sensor 17b can recognize the white line.

  Information indicating the shape of the host vehicle travel lane, the positional relationship between the host vehicle travel lane and the host vehicle in the lane width direction, and the like may be provided from the navigation ECU 90.

  The operation switch 18 shown in FIG. 1 is a switch operated by the driver of the host vehicle SV. The driver can select whether to execute lane keeping control including steering follow-up control described later by operating the operation switch 18. Furthermore, the driver can select whether to execute an inter-vehicle distance control (following inter-vehicle distance control) described later by operating the operation switch 18.

  The yaw rate sensor 19 detects a yaw rate of the host vehicle SV and outputs an actual yaw rate YRt.

  The engine ECU 30 is connected to the engine actuator 31. The engine actuator 31 is an actuator for changing the operating state of the internal combustion engine 32, and includes at least a throttle valve actuator for changing the opening degree of the throttle valve. Engine ECU 30 can change the torque generated by internal combustion engine 32 by driving engine actuator 31, thereby controlling the driving force of host vehicle SV to change the acceleration of host vehicle SV. it can.

  The brake ECU 40 is connected to the brake actuator 41. The brake actuator 41 is provided in a hydraulic circuit between a master cylinder (not shown) and the friction brake mechanism 42 provided on the left and right front and rear wheels. The brake actuator 41 adjusts the hydraulic pressure of the hydraulic fluid supplied to the wheel cylinder built in the brake caliper 42b of the friction brake mechanism 42 according to the instruction from the brake ECU 40, and the hydraulic pressure of the hydraulic fluid supplied to the wheel cylinder is not shown. To generate friction braking force. Accordingly, by controlling the brake actuator 41, the brake ECU 40 can control the braking force of the host vehicle SV to change the acceleration (in this case, the deceleration) of the host vehicle SV.

  The steering ECU 60 is a control device of a known electric power steering system, and is connected to the motor driver 61. The motor driver 61 is connected to the steering motor 62. The steering motor 62 is incorporated in a "steering mechanism including a steering wheel SW, a steering shaft US, a steering gear mechanism (not shown) and the like". The steering motor 62 generates a torque by the power supplied from the motor driver 61, and can generate a steering assist torque or steer the left and right steered wheels by this torque. That is, the steering motor 62 can change the steering angle (also referred to as “steering angle” or “steering angle”) of the host vehicle SV.

  The meter ECU 70 is connected to a digital display meter not shown. Furthermore, the meter ECU 70 is also connected to the hazard lamp 71 and the stop lamp 72, and can change the lighting state of these according to an instruction from the DSECU.

  The alarm ECU 80 is connected to the buzzer 81 and the display 82. The alarm ECU 80 can alert the driver by ringing the buzzer 81 in response to an instruction from the DSECU, and makes the indicator 82 light a mark for alerting (for example, a warning lamp) can do.

  The navigation ECU 90 is connected to a GPS receiver 91 for receiving a GPS signal for detecting the current position of the vehicle SV, a map database 92 storing map information and the like, a touch panel display 93 and the like. The navigation ECU 90 determines the position of the vehicle SV at the current time based on the GPS signal (in the case where the vehicle SV is traveling on a road having a plurality of lanes, which lane the vehicle SV is traveling) Identify information). The navigation ECU 90 performs various arithmetic processing based on the position of the vehicle SV and the map information and the like stored in the map database 92, and performs route guidance using the display 93 based on the arithmetic processing result.

<Overview of operation>
Next, an outline of the operation of the present embodiment will be described. The DSECU of this embodiment can execute inter-vehicle distance control and lane keeping control. Hereinafter, “inter-vehicle distance control and lane keeping control” will be described.

<Inter-vehicle distance control (ACC: adaptive cruise control)>
The inter-vehicle distance control (that is, the following inter-vehicle distance control) is an inter-vehicle distance between the own vehicle SV and a forward vehicle traveling in front of the own vehicle SV in the area ahead of the own vehicle SV based on target information. (That is, control is performed to cause the own vehicle SV to follow the preceding vehicle while maintaining the longitudinal distance Dfx (n) of the preceding vehicle with respect to the own vehicle SV at a predetermined target inter-vehicle distance. The following inter-vehicle distance control itself is known (see, for example, JP-A-2014-148293, JP-A-2006-315491, JP-A-4172434, and JP-A-4929777). Therefore, it will be briefly described below.

  The DSECU executes inter-vehicle distance control when inter-vehicle distance control is requested by the operation of the operation switch 18.

  First, when the inter-vehicle distance control is requested, the DSECU calls a vehicle to be followed based on the target information of the target (n) acquired by the surrounding sensor 17 (hereinafter referred to as “inter-vehicle distance target vehicle” Identified). More specifically, the DSECU determines (specifies) an inter-vehicle distance target vehicle among other vehicles (that is, forward vehicles) traveling in the front region of the host vehicle SV as follows.

Step 1A: The DSECU acquires, from the vehicle speed sensor 16 and the yaw rate sensor 19, the vehicle speed SPD of the vehicle SV and the yaw rate Yrt of the vehicle SV, which are motion state quantities of the vehicle SV.
Step 2A: The DSECU predicts the traveling path of the host vehicle SV in xy coordinates based on the vehicle speed SPD and the yaw rate Yrt.
Step 3A: The DSECU determines that the predicted absolute value of the distance in the lane width direction from the travel route of the own vehicle SV predicted from among other vehicles (ie, forward vehicles) having a positive value of the vertical distance Dfx (n) The other vehicle within the first reference threshold of is determined (selected / set) as an inter-vehicle distance target vehicle (a). The first reference threshold is set to decrease as the vertical distance Dfx (n) increases. When there are a plurality of other vehicles determined, the DSECU specifies the other vehicle with the minimum vertical distance Dfx (n) as the target inter-vehicle distance (a).

  When the inter-vehicle distance target vehicle (a) is specified, the DSECU calculates the target acceleration Gtgt in accordance with either of the following equations (1) and (2). In the equations (1) and (2), Vfx (a) is the relative velocity of the target inter-vehicle distance (a), k1 and k2 are predetermined positive gains (coefficients), and ΔD1 is the target of inter-vehicle distance An inter-vehicle deviation (ΔD1 = Dfx (a) −Dtgt) obtained by subtracting the “target inter-vehicle distance Dtgt” from the longitudinal distance Dfx (a) of the vehicle (a). The target inter-vehicle distance Dtgt is calculated by multiplying the target inter-vehicle time Ttgt set by the driver using the operation switch 18 by the vehicle speed SPD of the host vehicle SV (that is, Dtgt = Ttgt · SPD).

When the value (k1 · ΔD1 + k2 · Vfx (a)) is positive or “0”, the DSECU determines the target acceleration Gtgt using the following equation (1). ka1 is a positive gain (coefficient) for acceleration, and is set to a value of “1” or less.
When the value (k1 · ΔD1 + k2 · Vfx (a)) is negative, the DSECU determines the target acceleration Gtgt using the following equation (2). kd1 is a positive gain (coefficient) for deceleration, and is set to "1" in this example.

Gtgt (for acceleration) = ka1 · (k1 · ΔD1 + k2 · Vfx (a)) (1)
Gtgt (for deceleration) = kd1 · (k1 · ΔD1 + k2 · Vfx (a)) (2)

  When the target vehicle distance can not be specified due to the absence of the preceding vehicle, the DSECU is configured such that the vehicle speed SPD of the host vehicle SV matches the "target vehicle speed set using the operation switch 18". Then, the target acceleration Gtgt is determined based on the target vehicle speed and the vehicle speed SPD.

  The DSECU controls the engine actuator 31 using the engine ECU 30 so that the acceleration of the host vehicle SV matches the target acceleration Gtgt, and controls the brake actuator 41 using the brake ECU 40 as necessary.

<Lane maintenance control>
When the lane keeping control is requested by the operation of the operation switch 18, the DSECU executes the lane keeping control only during the execution of the inter-vehicle distance control. Lane maintenance control mainly includes section lane maintenance control and steering follow-up control.

  Section lane maintenance control determines a target travel line (target travel path) based on the division lines such as the white line and the yellow line, and the steering angle of the host vehicle SV so that the host vehicle SV travels along the target travel line. Control to adjust the Partition lane maintenance control may be referred to as LTC (Lane Trace Control). In the following, the dividing lines are described as white lines.

  The steering follow-up control identifies one of the preceding vehicles as the steering follow-up target vehicle TV, and the own vehicle SV travels along the target traveling line according to the traveling track of the steering follow-up target vehicle TV. It is control which adjusts a steering angle (refer FIG. 2). Steering follow-up control and section lane maintenance control may also be generically referred to as "TJA (Traffic Jam Assist)", and may be referred to as "steering support control" because the control assists the driver's steering operation. . Hereinafter, the description will be added in the order of section lane maintenance control and then steering follow-up control.

<< Section Lane Maintenance Control >>
If at least one of the left white line and the right white line is recognized by the camera sensor 17b over a predetermined distance in the forward direction of the host vehicle SV, the DSECU determines a target travel line based on at least one of the left white line and the right white line. Set Ld. Note that the DSECU acquires the lane width (that is, the distance between the left white line and the right white line) based on the recognized “position of left white line and right white line” every time a predetermined time elapses, and stores it in the RAM as needed. ing.

  More specifically, when both the left white line and the right white line are recognized in the forward direction of the host vehicle SV over a predetermined distance, the DSECU measures the center position in the lane width direction of the left white line and the right white line. The line passing through (ie, the center line) is set as the target travel line Ld.

  On the other hand, when only one white line of the left white line and the right white line is recognized over the predetermined distance in the forward direction of the host vehicle SV, the DSECU recognizes the recognized white line and the left white line. The position of the unrecognized white line (the other white line) is estimated based on the lane width acquired when both the right white line and the right white line were recognized. Then, the DSECU sets the center line of the recognized one white line and the estimated other white line as the target travel line Ld.

  Furthermore, the steering motor is controlled so that the DSECU maintains the lateral position of the host vehicle SV (that is, the position of the host vehicle SV in the lane width direction with respect to the host vehicle travel lane) in the vicinity of the set target travel line Ld. The steering angle of the host vehicle SV is changed by applying a steering torque to the steering mechanism using 62, thereby supporting the driver's steering operation (for example, Japanese Patent Application Laid-Open Nos. 2008-195402 and 2009-). 190464, JP-A-2010-6279, and Patent No. 4349210, etc.). A specific steering control method will be described later.

<< Steering tracking control >>
If there is no white line recognized over a predetermined distance in the forward direction of the host vehicle SV, the DSECU is suitable as a steering-following target vehicle TV from among other vehicles (forward vehicles) traveling in the front region of the host vehicle SV. Select a vehicle ahead. Then, the DSECU generates a traveling locus of the steering follow-up target vehicle TV (hereinafter also referred to as "preceding vehicle locus"), and the host vehicle SV travels along a target traveling line determined based on the preceding vehicle locus. The steering torque is applied to the steering mechanism to change the steering angle. In the present example, the DSECU sets the preceding vehicle trajectory itself as the target travel line Ld. However, the DSECU may set a line displaced in the lane width direction by a predetermined distance from the preceding vehicle trajectory as the target travel line Ld.

  Next, the method of determining the steering follow-up target vehicle TV, the method of generating the preceding vehicle trajectory, and the method of steering follow-up control will be added.

1. Method of Determining a Steering Follow-Up Target Vehicle As shown in FIG. 2, the DSECU sets a forward vehicle, which is a target (n) for which a preceding vehicle trajectory is to be created, as a steering follow-up target vehicle TV. The method of setting the steering following target vehicle TV will be described in detail later.

2. Generation of Preceding Vehicle Trajectory As shown in FIG. 2, the DSECU generates a traveling trajectory L1 (that is, a preceding vehicle trajectory) of the steering following target vehicle TV. More specifically, as shown in FIG. 3A, the traveling locus L1 is a table of cubic functions of the following equation (3) at the above-mentioned x-y coordinates at the current position of the vehicle SV: It is known to be accurately approximated by the

y = (1/6) Cv 'x ³ + (1/2) Cv x ² + θv x + dv (3)

Cv ′: curvature change rate (curvature change amount per unit distance (Δx) at an arbitrary position on the curve (x = x0, x0 is an arbitrary value)).
Cv: When the steering following target vehicle TV is present at the current position (x = 0) of the host vehicle SV (ie, the steering following target vehicle TV is present at the position (x = 0, y = dv) Curvature of the traveling locus L1).
θv: Direction of the traveling locus L1 (the tangential direction of the traveling locus L1) when the steering following target vehicle TV exists at the current position (x = 0) of the own vehicle SV and the traveling direction of the own vehicle SV (x axis Angular deviation with +). The angular deviation θv is also referred to as “yaw angle”.
dv: a distance dv (substantially in the lane width direction) between the current position (x = 0, y = 0) of the vehicle SV and the traveling locus L1. This distance dv is also referred to as "center distance".

The above equation (3) is derived as described below. That is, as shown in FIG. 3 (B), the traveling locus L1 is set to a cubic function f (x) = ax ³ + bx ² + cx + d, and further, the relational expression shown in FIG. 3 (B) and By using the conditions, “the relationship between the coefficients (a, b, c and d) of the cubic function and the curvature etc.” shown in FIG. 3C can be derived. Therefore, when the coefficients (a, b, c and d) of the cubic function are obtained from the relationship shown in FIG. 3C, the above equation (3) is derived.

  Based on this point of view, the DSECU finds the coefficients of the first to fourth terms of the right side of equation (3) (ie, the coefficients a, b, c, and d of the function f (x) as follows: .

  -The DSECU acquires target information of the steering following target vehicle TV (hereinafter may be referred to as "target (b)") every time a predetermined measurement time elapses, and the target information is obtained. Positional coordinate data representing the position (longitudinal distance Dfx (n) and lateral position Dfy (n)) of the steering following target vehicle TV at the time of acquisition is stored (buffered) in the RAM. In order to minimize the amount of data to be stored, the DSECU saves only a certain number of position coordinate data from the latest position coordinate data of the steering following target vehicle TV, and discards the old position coordinate data sequentially You may

  The DSECU stores the position coordinate data of the steering-following target vehicle TV stored in the RAM on the position and traveling direction of the vehicle SV, and the position and traveling direction of the vehicle SV at the current time when acquiring each position coordinate data. Based on the difference between the two points, and is converted into position coordinate data of xy coordinates with the current position as the origin (x = 0, y = 0). The x and y coordinates of the converted position coordinate data (hereinafter sometimes referred to as "converted position coordinates") are placed as xi and yi, respectively. In this case, (xi, yi) = (x1, y1), (x2, y2), (x3, y3), and (x4, y4) shown in FIG. It is an example of the position coordinates after conversion of TV.

  The DSECU executes a curve fitting process using the post-change position coordinates of the steering following target vehicle TV to generate a traveling locus L1 of the steering following target vehicle TV. The curve used for this fitting process is a cubic curve (curve represented by the above-described cubic function f (x)). The fitting process is performed by the least squares method.

3. Execution of Steering Follow-up Control The DSECU sets the generated traveling locus L1 to the target traveling line Ld. Furthermore, the DSECU is required for steering following control when the traveling locus L1 is set to the target traveling line Ld based on the coefficient of the cubic function of the equation (3) and the relational expression shown in FIG. Information (hereinafter, sometimes referred to as "target travel path information") is acquired. The target travel path information includes the curvature Cv of the travel locus L1, the yaw angle θv with respect to the travel locus L1, and the center distance dv with respect to the travel locus L1.

The DSECU calculates the target steering angle θ * by applying the curvature Cv, the yaw angle θv and the center distance dv to the following equation (4) each time a predetermined time has elapsed. In the equation (4), Klta1, Klta2 and Klta3 are predetermined control gains. Furthermore, the DSECU controls the steering motor 62 using the steering ECU 60 so that the actual steering angle θ matches the target steering angle θ *. By the above, steering control by steering following control is performed.

θ * = Klta1 · Cv + Klta2 · θv + Klta3 · dv (4)

Note that the DSECU may calculate the target yaw rate YRc * by applying the curvature Cv, the yaw angle θv, and the center distance dv to the following equation (5) each time a predetermined time elapses. In this case, the DSECU calculates a target steering torque Tr * for obtaining the target yaw rate YRc * using the look-up table, based on the target yaw rate YRc * and the actual yaw rate YRt. Then, the DSECU controls the steering motor 62 using the steering ECU 60 so that the actual steering torque Tra matches the target steering torque Tr *. Also by the above, steering control by steering following control is performed. As understood from the above, if the DSECU can acquire the target travel path information, it can execute the steering follow-up control without calculating the target travel line itself.

YRc * = K1 × dv + K2 × θv + K3 × Cv (5)

  The DSECU also uses the above equation (4) or (5) when executing the above-described section lane maintenance control. More specifically, the DSECU sets the curvature CL of the target travel line Ld (that is, the center line of the host vehicle travel lane) set based on at least one of the left white line and the right white line, and the vehicle width of the host vehicle SV. The distance dL between the central position in the direction and the target travel line Ld in the y-axis direction (substantially the road width direction) and the deviation between the direction of the target travel line Ld (tangential direction) and the traveling direction of the host vehicle SV The angle θL (yaw angle θL) is calculated.

  Then, the DSECU calculates the target steering angle θ * by replacing dv with dL, replacing θv with θL, and replacing Cv with CL in equation (4) (or equation (5)). The steering motor 62 is controlled such that the actual steering angle θ matches the target steering angle θ *. Thus, the steering control based on the block lane maintenance control is performed.

  The DSECU can not set the target travel line Ld based on at least one of the left white line and the right white line, and can not generate the preceding vehicle locus (including the case where the steering follow-up target vehicle TV can not be determined). Cancel execution of lane keeping control. That is, in this case, the DSECU does not perform lane keeping control.

  Next, referring to FIGS. 4A and 4B, and FIGS. 5 to 8, the DSECU of the present embodiment executes “the situation where the target vehicle TV following the steering deviates from the traveling lane. The method for determining whether or not the

  As shown in FIG. 4A, the DSECU is currently performing steering control (steering follow control) so that the host vehicle SV follows the travel locus L1 of the target vehicle TV.

  The DSECU detects the steering following target vehicle TV (that is, the past steering following target vehicle TVp) and the white line LW at the time when the DSECU exists at a position on the traveling locus L1 corresponding to the position of the own vehicle SV The first parameters (the first distance DL1 and the first yaw angle θ1 (see FIG. 5)) obtained by estimating the positional relationship with are obtained as follows.

(Acquisition of the third distance DL3 and the third yaw angle θ3)
First, as shown in FIG. 4A, the DSECU acquires a third distance DL3 in the lane width direction between the host vehicle SV and the host vehicle corresponding position. Furthermore, the DSECU acquires a third yaw angle θ3 with respect to the travel locus L1 of the steering following target vehicle TV of the host vehicle SV.

  The "vehicle-corresponding position" is a position where the vertical distance is the same as the vertical distance of the vehicle SV, and the horizontal position is on the traveling locus L1. The lateral position of the host vehicle corresponding position is a value that can be obtained by substituting the vertical distance of the host vehicle SV for the variable x of the equation (3) representing the traveling locus L1.

(Acquisition of the fourth distance DL4 and the fourth yaw angle θ4)
Next, as shown in FIG. 4B, a fourth distance DL4 in the lane width direction between the vehicle SV and the white line LW is acquired. Furthermore, the DSECU acquires a fourth yaw angle θ4 with respect to the white line LW of the host vehicle SV.

(Acquisition of the first distance DL1 and the first yaw angle θ1)
Then, as shown in FIG. 5, the DSECU obtains the first distance DL1 by subtracting the third distance DL3 from the fourth distance DL4 (first distance DL1 = fourth distance DL4-third distance DL3). Do.

  Furthermore, the DSECU calculates the difference between the third yaw angle θ3 and the fourth yaw angle θ4 (“first yaw angle θ1” = “third yaw angle θ3” − “fourth yaw angle θ4”). 1 Acquire the yaw angle θ1. Thus, the first parameter (the first distance DL1 and the first yaw angle θ1) is acquired.

  Furthermore, as shown in FIG. 6, the DSECU estimates the first parameter (the first distance DL1 and the first yaw angle θ1) as an observation value to the Kalman filter 10a included in the DSECU as an output value by estimating it as an output value. The first distance DL1f, the estimated first yaw angle θ1f, and the first yaw angle change rate θ1 ′ are acquired.

  Specifically, the Kalman filter 10a performs Kalman filter processing using (the first distance DL1 and the first yaw angle θ1) as the observed value, σ1 as the process noise, σ2 as the observation noise, and σ3 as input values. A first distance DL1 (estimated first distance DL1f), a first yaw angle θ1 (estimated first yaw angle θ1f), and a first yaw angle change rate θ1 ′ are calculated as later output values. .

  The Kalman filter 10a (extended Kalman filter), as is well known, describes a nonlinear dynamic system, taking into account the state equation of the same system and the observation equation under the influence of noise, and based on those equations It is an iterative estimation type filter that estimates the current state quantity from the observed value of the current state and the estimated value of the past state quantity (for example, JP-A-2007-505377, JP-A-2014-10872, JP-A-2014) -102137, JP-A 2014-2103 and JP-A 2017-012650 etc.)

  The Kalman filter 10a is an observation equation having (DL1, .theta.1) as observation value and (.sigma.2, .sigma.3) as observation noise, and a state equation having state quantities (DL1, .theta.1, .theta.1 ') and process noise (.sigma.1) To perform the well-known karma filter processing according to the well-known Kalman filter principle. Thus, the DSECU can acquire the estimated first distance DL1f, the estimated first yaw angle θ1f, and the first yaw angle change rate θ1 '.

(Calculation of σ1 to σ3)
The DSECU calculates σ 1 to σ 3 input to the Kalman filter 10 a as follows as follows.

(Operation of σ1)
σ1 is “a standard deviation σΔθ1 ′ of the change amount Δθ1 ′ of the first yaw angle change rate θ1 ′ per operation cycle (predetermined time)”. Note that σ 1 is also referred to as “first standard deviation” for the sake of convenience.

  The DSECU calculates the vehicle speed v of the steering following target vehicle TV, the lane width W, the maximum lateral speed Vmax, and the calculation cycle ΔTcy by substituting the following (calculation formula 1). An experimentally obtained value is set as the maximum lateral velocity Vxmax. As the maximum lateral velocity Vxmax, for example, the maximum lateral velocity (fixed value) under the assumption that the vehicle subject to the steering follow-up has made a sudden lane change is set. Note that the maximum lateral speed (a value that varies according to the vehicle speed v of the steering following target vehicle TV) estimated based on the detected vehicle speed v of the steering following target vehicle TV may be set as the maximum lateral speed Vxmax. .

(Calculation formula 1)
σ1 (= σΔθ1 ′) = (4Vxmax ³ / W ² ) × (Δtcy / 3v)
(Σ1: “Standard deviation of change amount of first yaw angle change rate θ1 ′ per calculation cycle (σΔθ1 ′)”, Vxmax: maximum lateral velocity, W: lane width, Δtcy: calculation cycle, v: steering following target vehicle Speed of

  (Formula 1) is a formula derived as follows. It is assumed that the steering-following target vehicle TV has made a sudden lane change. At this time, when the maximum lateral velocity Vxmax of the steering following target vehicle TV and the time from the lane change start to the completion are Tf, the relationship between the lateral velocity Vx of the steering following target vehicle TV and the time is shown in FIG. It becomes as shown in graph 1.

  During the time from 0 to 0.5 Tf, the lateral acceleration of the line m1 indicating the lateral speed of the steering following target vehicle TV is constant as shown by the line m2, and becomes Vxmax / 0.5 Tf. In the range of time from 0 to 0.5 Tf, the lateral acceleration change indicated by the line m2 is indicated by the line m3 having the same area as the area representing the speed of the lateral acceleration change Converted into lateral acceleration change.

  That is, as shown by the line m3, the maximum lateral acceleration is obtained with the lateral jerk Jx (the amount of change in lateral acceleration per unit time) in which the lateral acceleration of the steering following target vehicle is constant during the time from 0 to 0.25Tf. After rising up, it is converted into a change in lateral acceleration which decreases to 0 at a constant lateral jerk Jx from 0.25 Tf to 0.5.

At this time, the horizontal jerk Jx (hereinafter referred to as “maximum horizontal jerk Jxmax”) is (2Vxmax ÷ 0.5Tf) ÷ 0.25T = (16Vxmax / Tf ² ) (= maximum horizontal jerk Jxmax). -It becomes (1a). The lane width W can be calculated by time-integrating the function indicated by the line m1 of the graph 1 in the range from time 0 to Tf. That is, the lane width W = 1/2 × Tf × Vxmax (that is, Tf = 2W / Vxmax). Then, when Tf = 2 W / Vxmax is substituted into the above (1a), the maximum lateral jerk Jxmax = 4Vxmax ³ / W ² is obtained.

Here, assuming that the maximum lateral jerk Jxmax is a value with an occurrence probability of 0.3% according to a normal distribution with an average of 0, the standard deviation of the lateral jerk is 3σjx = 4Vxmax ³ / W ² . The relational expression (σjx = (v / Δtcy) σΔθ1 ′) between “standard deviation σjx of the lateral jerk” and “standard deviation σΔθ1 ′ of the change amount of the first yaw angle change rate θ1 ′ per operation cycle” is expressed by the above equation When substituted, σΔθ1 ′ = (4Vxmax ³ / W ² ) × (Δtcy / 3v) is derived. By the above, it is possible to derive (Expression 1), and the standard deviation of “the first yaw angle change rate θ1 ′ change amount Δθ1 ′ according to the vehicle speed v of the steering following target vehicle TV and the lane width W of the traveling lane. It is possible to estimate σΔθ1 ′ (ie, σ1).

(Operation of σ2 and σ3)
σ2 is “standard deviation of error of first distance DL1”. The DSECU calculates σ2 using the following (calculation formula 2). The symbol σ3 is “a standard deviation of the error of the first yaw angle θ1”. The DSECU calculates the symbol σ3 using the following (calculation formula 3). Note that σ 2 is also referred to as “second standard deviation” for the sake of convenience. σ 3 is also referred to as “third standard deviation” for convenience.

(Calculation formula 2)
σ 2 = ((“standard deviation of the error of the third distance DL3”) ² + (“standard deviation of the error of the fourth distance DL4”) ² ) ^0.5
(Calculation formula 3)
σ3 = ((“standard deviation of error of the third yaw angle θ3”) ² + (“standard deviation of error of the fourth yaw angle θ4”) ² ) ^0.5

  For each of the “standard deviation of the error of the third distance DL3” and the “standard deviation of the error of the third yaw angle θ3”, for example, experimentally obtained values are appropriately set. The “standard deviation of the fourth distance DL4 error” and the “standard deviation of the error of the fourth yaw angle θ4” are set to different values according to the accuracy of the white line recognition.

  Specifically, when the white line recognition accuracy is high, the DSECU sets the “standard deviation of the fourth distance DL4 error” to the first value. If the white line recognition accuracy is low, the DSECU sets the “standard deviation of the fourth distance DL4 error” to a second value larger than the first value. Note that, for example, experimentally obtained values are appropriately set as the first value and the second value.

  Furthermore, if the white line recognition accuracy is high, the DSECU sets the “standard deviation of the fourth yaw angle θ4 error” to a third value. If the white line recognition accuracy is low, the DSECU sets “the standard deviation of the fourth yaw angle θ4 error” to a fourth value larger than the third value. As the third value and the fourth value, experimentally obtained values are appropriately set.

(Judgment method)
Furthermore, as shown in FIG. 8, the DSECU is configured to calculate a first parameter (a first distance DL1 (DL1 f) as an output value after Kalman filter processing, a first yaw angle θ1 (θ1 f), and a first yaw angle change rate θ1 ′. ), The vehicle speed v of the steering following target vehicle TV, and the first moving time Tx2 taken for the steering following target vehicle TV to move from the position corresponding to the host vehicle to the current position (= “vehicle distance / steering following target vehicle TV v) etc., the second yaw angle θ2 which is the yaw angle with respect to the white line LW of the steering following target vehicle TV, and the steering following target vehicle TV and the white line according to the following (first formula) to (third formula) The second parameter (decision parameter) below is obtained by estimating the positional relationship with LW.

(Judgment parameter)
Second distance DL2: Distance in the lane width direction between the steering following target vehicle TV and the white line LW Estimated lateral speed v2: Estimated lateral speed of the steering following target vehicle TV Ta: Departing prediction time Ta: Departing prediction time Ta = Prediction time until the steering following target vehicle TV calculated at the second distance DL2 / estimated lateral speed v2 reaches the white line LW

（ＤＬ２：第２距離、ＤＬ１：第１距離、ｖ：操舵追従目標車両の車速、θ１：第１ヨー角、θ１’：第１ヨー角変化率、Ｔｘ２：第１移動時間）

θ2＝θ1+(θ１’×Ｔｘ２)・・・（第２数式）

（θ２：第２ヨー角、θ１：第１ヨー角、θ１’：第１ヨー角変化率、Ｔｘ２：第１移動時間）

ｖ２＝ｖ×θ２・・・（第３数式）
（ｖ２：推定横速度、ｖ：操舵追従目標車両の速度、θ２：第２ヨー角）

(DL2: second distance, DL1: first distance, v: vehicle speed of steering target vehicle, θ1: first yaw angle, θ1 ′: first yaw angle change rate, Tx2: first movement time)

θ2 = θ1 + (θ1 ′ × Tx2)... (second equation)

(Θ2: second yaw angle, θ1: first yaw angle, θ1 ′: first yaw angle change rate, Tx2: first movement time)

v2 = v × θ2 (third equation)
(V2: estimated lateral speed, v: speed of target vehicle following with steering, θ2: second yaw angle)

  The DSECU uses the acquired determination parameters (the second distance DL2, the estimated lateral velocity v2, and the departure prediction time Ta) to determine the departure situation of the steering following target vehicle TV from the travel lane.

  Specifically, the DSECU first determines, using the determination parameter, whether or not the steering-following target vehicle TV is in a departure situation deviating from the traveling lane.

If at least one of the following departure situation determination conditions 1 to 3 is satisfied, the DSECU determines that the steering following target vehicle TV is in a departure situation.
Deviation situation determination condition 1: The second distance DL2 is equal to or less than the first threshold distance Dth1.
Deviation Situation Determination Condition 2: The estimated lateral velocity v2 is equal to or greater than the first threshold lateral velocity vth1.
Deviation situation determination condition 3: The departure prediction time Ta (= DL2 / v2) is equal to or less than the first threshold time Tth1.

  If the steering following target vehicle TV is in a departure situation, the DSECU cancels the steering following control. If the steering follow-up target vehicle TV is not in a departure situation, the DSECU performs a steering follow-up control.

<Concrete operation>
Next, the specific operation of the CPU of the DSECU (which may be simply referred to as “CPU”) will be described. The CPU is configured to execute a steering follow-up control routine shown by the flowchart of FIG. 9 each time a predetermined time (calculation cycle Δtcy) elapses. The CPU executes inter-vehicle distance control (ACC) by a routine not shown. The CPU executes the routine shown in FIG. 9 only when the inter-vehicle distance control is being performed.

  Therefore, when the inter-vehicle distance control is being executed, the CPU starts the process from step 900 in FIG. 9 at a predetermined timing and proceeds to step 905 to determine whether the steering follow-up control execution condition is satisfied. Determine if

The execution condition of the steering follow-up control is satisfied, for example, when all of the conditions A1 to A3 described below are satisfied.
Condition A1: It is selected by the operation of the operation switch 18 to execute lane keeping control.
Condition A2: The vehicle speed SPD of the host vehicle SV is equal to or higher than a predetermined lower limit vehicle speed and equal to or lower than a predetermined upper limit vehicle speed.
Condition A3: The target travel line Ld based on "at least one of the left white line and the right white line" recognized by the camera sensor 17b can not be set.

  If the execution condition of the steering follow-up control is not satisfied, the CPU determines “No” in step 905, proceeds to step 910, and cancels (cancels) the steering follow-up control. Thereafter, the CPU proceeds to step 995 to end this routine once.

  On the other hand, when the execution condition of the steering follow-up control is satisfied, the CPU makes a “Yes” determination in step 905, proceeds to step 915, and performs the steering follow-up target vehicle TV for which the travel locus L1 is generated. Identify. Specifically, the CPU acquires the vehicle speed of the host vehicle SV from the vehicle speed sensor 16 and acquires the yaw rate of the host vehicle SV from the yaw rate sensor 19. The CPU predicts the traveling route of the host vehicle SV from the acquired vehicle speed and yaw rate. Then, the target closest to the preceding vehicle closest to the predicted "traveling path of the host vehicle SV" is selected as the "steering-following target vehicle TV to be generated as the traveling locus L1".

  The CPU stores target information of each target in correspondence with each target based on the target information from the surrounding sensor 17. The CPU selects the target information for the steering following target vehicle TV specified from the target information, and generates a traveling locus L1 for the steering following target vehicle TV based on the selected target information.

  Thereafter, the CPU proceeds to step 920 and determines whether or not the travel locus L1 can be generated. Specifically, when the steering following target vehicle TV can not be specified, or the steering following target vehicle TV can be specified, the time-series data of the target information about the steering following target vehicle TV is the traveling locus If it is not sufficient to generate L1, the CPU determines that the traveling locus L1 can not be generated. Otherwise, the CPU determines that the traveling locus L1 can be generated.

  If the traveling locus L1 can be generated, the CPU makes a “Yes” determination in step 920, proceeds to step 925, and generates a white line by the camera sensor 17b based on the “recognition distance information” sent from the camera sensor 17b. It is determined whether or not the left white line and the right white line can be recognized in a range not less than the first predetermined distance and less than the second predetermined distance. In other words, the CPU determines whether the white line can be recognized in the vicinity by the camera sensor 17b. The first predetermined distance is set smaller than the second predetermined distance.

  If the white line is recognized by the camera sensor 17b in the range from the first predetermined distance to the second predetermined distance, the CPU determines “Yes” in step 925, proceeds to step 930, and proceeds from step 930 to steps described below. After the process of 960 is sequentially performed, the process proceeds to step 965.

Step 930: The CPU acquires a third distance DL3 in the lane width direction between the host vehicle SV and the host vehicle corresponding position. Further, the CPU acquires a third yaw angle θ3 with respect to the travel locus L1 of the steering following target vehicle TV of the host vehicle SV.
Step 935: The CPU obtains a fourth distance DL4 in the lane width direction between the host vehicle SV and the white line LW. Furthermore, the CPU obtains a fourth yaw angle θ4 with respect to the white line LW of the host vehicle SV.
Step 940: The CPU obtains the first distance DL1 by subtracting the third distance DL3 from the fourth distance DL4 (first distance DL1 = fourth distance DL4−third distance DL3).
The CPU obtains the first yaw angle θ1 by calculating the difference between the third yaw angle θ3 and the fourth yaw angle θ4 (first yaw angle θ1 = third yaw angle θ3−fourth yaw angle θ4). .

Step 945: The CPU calculates (estimates) σ1 using (calculation formula 1) as described above.
Step 950: The CPU calculates (estimates) σ2 using (Formula 2) as described above. At this time, if the white line recognition accuracy is high, the CPU sets the “fourth distance DL4 error standard deviation” of (Expression 2) to a first value. When the white line recognition accuracy is low, the CPU sets “the standard deviation of the fourth distance DL4 error” of (Expression 2) to a second value larger than the first value.
The CPU calculates σ 3 using (calculation formula 3) as described above. At this time, if the white line recognition accuracy is high, the CPU sets “the standard deviation of the fourth yaw angle θ4 error” of (Expression 3) to a third value. If the white line recognition accuracy is low, the CPU sets “the standard deviation of the fourth yaw angle θ4 error” of (Expression 3) to a fourth value larger than the third value.

  Step 955: The CPU obtains the estimated first distance DL1f, the estimated first yaw angle θ1f, and the first yaw angle change rate θ1 ′ using the Kalman filter 10a as described above.

  Step 960: The CPU uses the estimated first distance DL1f, the estimated first yaw angle θ1f, the first yaw angle change rate θ1 ′ (first parameter), and the vehicle speed v and the first movement time Tx2 as described above. Then, the determination parameters (the second distance DL2, the estimated lateral velocity v2, and the deviation prediction time Ta) are calculated.

  When the CPU proceeds to step 965, it determines, based on the determination parameter, whether the steering following target vehicle TV is in a departure situation. The CPU determines that the steering following target vehicle TV is in a departure situation if at least one of the departure situation determination conditions 1 to 3 described above is satisfied.

  If the steering follow-up target vehicle TV is not in a departure situation, the CPU determines "No" in step 965, proceeds to step 970, and proceeds to step 915 to generate a traveling locus L1 generated based on the steering follow-up target vehicle as a target traveling line. The steering angle of the host vehicle SV is controlled (steering control is performed) so that the host vehicle SV is set to Ld and the host vehicle SV is caused to travel along the target travel line Ld. That is, the CPU executes steering follow-up control. Thereafter, the CPU proceeds to step 995 to end this routine once.

  If the steering follow-up target vehicle TV is in a departure situation, the CPU makes a “Yes” determination in step 965 and proceeds to step 910 to cancel (cancel) the steering follow-up control. Thereafter, the CPU proceeds to step 995 to end this routine once.

  The present embodiment described above has the following effects. That is, according to the vehicle speed v of the steering following target vehicle TV and the lane width W of the traveling lane, the value of σ1 input as the process noise to the Kalman filter 10a is determined. Furthermore, in accordance with the recognition accuracy of the white line, the values of σ2 and σ3 input as the observation noise are determined. As a result, the present apparatus can accurately acquire the first yaw angle change rate θ1 ′ used for calculating the determination parameter, and can improve the accuracy of the determination parameter. As a result, it is possible to reduce the possibility that the determination accuracy as to whether or not the steering following target person vehicle deviates from the traveling lane.

<Modification>
As mentioned above, although embodiment of this invention was described concretely, this invention is not limited to the above-mentioned embodiment, Various deformation | transformation based on the technical idea of this invention are possible.

  For example, the present embodiment is configured to execute the lane keeping control only during execution of the following inter-vehicle distance control, but is configured to execute the lane maintaining control even when the following inter-vehicle distance control is not being executed. May be

  For example, the present implementation device may acquire position information, speed information, and the like of another vehicle including the steering follow-up target vehicle TV and the inter-vehicle distance target vehicle through inter-vehicle communication. Specifically, for example, the position information of the other vehicle acquired by the other vehicle by the navigation device of the other vehicle is transmitted to the own vehicle SV together with a vehicle ID signal specifying the other vehicle itself, and the own vehicle SV Based on the transmitted information, position information of the steering following target vehicle TV and / or the inter-vehicle distance target vehicle may be acquired.

  For example, in the present embodiment, the method of generating the traveling locus is not limited to the above-described example, and various known methods can be adopted. That is, any method may be used as long as it can create a curve that approximates the traveling locus (preceding vehicle locus) of the steering following target vehicle. For example, the traveling locus may be generated using a Kalman filter.

  Specifically, in the step 915, the CPU may generate a travel locus using a Kalman filter. When the position information of the host vehicle is input to the Kalman filter included in the driving support ECU 10, the curvature Cv of the travel locus L1 of the steering following vehicle of the current position of the host vehicle SV, the curvature change rate Cv 'of the travel locus L1, and the travel locus L1. The yaw angle θv of the vehicle SV and the distance dv between the traveling locus L1 and the current position of the vehicle SV are output. The CPU can obtain the coefficients a, b, c and d of the cubic function f (x) by using the relation between the coefficient of the cubic function and the curvature, the yaw angle, etc. shown in FIG. 3C. it can.

  DESCRIPTION OF SYMBOLS 10 ... Driving assistance ECU, 16 ... Vehicle speed sensor, 17 ... Ambient sensor, 17a ... Radar sensor, 17b ... Camera sensor, 17c ... Target object recognition part, 18 ... Operation switch, 19 ... Yaw rate sensor, 60 ... Steering ECU, 61 ... Motor driver, 62: Steering motor, 80: Alarm ECU, 81: Buzzer, 82: Display, SV: Self-vehicle, TV: Steering-following target vehicle, TVp: Past steering-following target vehicle

Claims (1)

A lane marking recognition unit that recognizes lane markings of a traveling lane in which the host vehicle is traveling;
A marking line recognition accuracy determination unit that determines the recognition accuracy of the marking lines;
A lane width acquisition unit that acquires the lane width of the traveling lane based on the recognized lane line;
A traveling locus generating unit that generates a traveling locus of a steering following target vehicle identified from among preceding vehicles traveling ahead of the host vehicle;
A vehicle speed acquisition unit that acquires the vehicle speed of the target vehicle following the steering;
The section of the past target vehicle following the steering when the longitudinal position is the same as the host vehicle and the lateral position exists at the host vehicle corresponding position at the position on the traveling locus every time the predetermined time elapses A first distance in a lane width direction between the line and the past steering following target vehicle, and a first yaw angle with respect to the division line of the past steering following target vehicle,
A third distance in the lane width direction between the host vehicle and the host vehicle corresponding position acquired based on the recognized dividing line and the travel locus, and the target vehicle of the target vehicle following the host vehicle Parameters acquired based on the third yaw angle with respect to the traveling track, the fourth distance in the lane width direction between the host vehicle and the lane line, and the fourth yaw angle of the host vehicle with respect to the lane line Acquisition part,
Substituting the vehicle speed of the target vehicle following the steering and the lane width into the following formula 1, the first standard deviation is calculated, and using the following formula 2, the standard deviation of the error of one distance which is the second standard deviation A standard deviation calculation unit that calculates the standard deviation of the error of the first yaw angle, which is the third standard deviation, using
When it is determined by the marking recognition accuracy determination unit that the marking recognition accuracy is high, the standard deviation of the error of the fourth distance in the calculation formula 2 is set to a first value, and in the calculation formula 3, Set the standard deviation of the 4th yaw angle error to 3
When it is determined by the marking recognition accuracy determination unit that the marking recognition accuracy is low, the standard deviation of the error of the fourth distance in the calculation formula 2 is set to the second value larger than the first value. The standard deviation calculation unit configured to set the standard deviation of the error of the fourth yaw angle in the formula 3 to a fourth value larger than the third value;
A parameter estimation unit equipped with a Kalman filter,
The obtained first distance and the first yaw angle are input to the Kalman filter as observation values each time the predetermined time passes.
Input the first standard deviation as process noise,
Input the second standard deviation and the third standard deviation as observation noise,
According to the principle of the Kalman filter, a first yaw angle change rate which is a change amount of the first yaw angle per predetermined time, an estimated first distance, and an estimated first yaw angle are calculated, and the calculated first yaw The parameter estimation unit configured to output an angular rate of change, an estimated first distance, and an estimated first yaw angle as an output value from the Kalman filter;
A first parameter including the first yaw angle change rate, the estimated first distance, and the estimated first yaw angle, and a second with respect to the division line of the steering following target vehicle calculated using the first parameter A second distance in the lane width direction between the steering following target vehicle and the lane line using at least one of the parameters including the yaw angle and the vehicle speed of the steering following target vehicle; Calculating an estimated lateral speed of the vehicle and an estimated time for the target vehicle following the steering to reach the lane line, based on at least one of the calculated second distance, estimated lateral speed, and estimated time,
A traveling lane departure situation determination unit that determines whether the steering-following target vehicle is in a departure situation that is departing from the traveling lane;
A traveling control unit that executes steering follow-up control that changes a steering angle of the own vehicle so that the own vehicle travels along a target traveling line set based on the traveling locus;
Equipped with
The traveling control unit stops the steering following control when it is determined that the steering following target vehicle is in the deviation situation.
The steering following control is executed when it is determined that the steering following target vehicle is not in the departure situation.
A vehicle driving support device configured as follows.
(Calculation formula 1)
1st standard deviation = (4Vxmax ³ / W ² ) x (Δtcy / 3v)
(First standard deviation: Standard deviation of the change amount of the first yaw angle change rate per predetermined time, Vxmax: maximum lateral velocity, W: the lane width, Δty: the predetermined time, v: the steering following target vehicle Speed of
(Calculation formula 2)
(Standard deviation of the first distance error) = ((standard deviation of the third distance error) ² + (standard deviation of the fourth distance error) ² ) ^0.5
(Calculation formula 3)
(Standard deviation of error of first yaw angle) = ((standard deviation of error of third yaw angle) ² + (standard deviation of error of fourth yaw angle) ² ) ^0.5

Images