JP2018127146A - Steering assist device for vehicle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent a vehicle from deviating from its own lane due to deviation of a steering angle neutral point.SOLUTION: A drive assist ECU 10 calculates a target steering angle θ* by adding a feed forward control term θFF, a feedback control term θFB, and an integral control term θI. The integral control term θI is an arithmetic term that is the integral of a lateral deviation Dy. A gain setting part 23a sets a lateral deviation integral control gain K4 at K4high in a preliminary-set specific period right after the start of LTA control. During the specific period, an own vehicle may deviate from its own lane due to deviation of a steering angle neutral point. The gain setting part 23a sets the lateral deviation integral control gain K4 at K4normal (<K4high) after a lapse of the specific period.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a vehicle steering assist device that supports a driver's steering operation so that the vehicle travels along a target travel line.

  2. Description of the Related Art Conventionally, as proposed in Patent Document 1, a vehicle steering assist device that assists a driver's steering operation so that the vehicle travels along a target travel line is known. In this steering assist device, the steering angle neutral point of the steering mechanism is learned using the detected value of the yaw rate sensor, and the steering angle is controlled with reference to the learned steering angle neutral point, so that the vehicle is moved along the target travel line. Let it run. As a result, the driver can receive assistance for the steering operation. Such control is called steering assist control.

JP 2010-120504 A

  In order to perform the steering assist control with high accuracy, it is necessary to accurately obtain the steering angle neutral point. However, the detection value of the yaw rate sensor changes due to its own temperature change. FIG. 4 shows detection characteristics of the yaw rate sensor. The horizontal axis represents time, the solid line represents the detection value of the yaw rate sensor, and the broken line represents the temperature of the yaw rate sensor. For example, immediately after the vehicle is started (ignition switch is turned on), the temperature of the yaw rate sensor rises due to the start of operation of the air conditioner, etc., so that the detection characteristics of the yaw rate sensor change.

  The steering angle neutral point is a steering angle when the vehicle travels straight. Therefore, if the detected value of the yaw rate sensor is accurate, it is possible to estimate straight traveling of the vehicle using the detected value, so that the required accuracy can be satisfied for the steering angle neutral point. However, before the temperature of the yaw rate sensor stabilizes, the detected value of the yaw rate sensor is not accurate, so if the detected value is used to estimate the vehicle's straight travel and find the steering angle neutral point, The rudder angle neutral point is not accurate.

  Generally, the steering angle neutral point is learned and stored during a period in which the ignition switch is on. Then, at the time of starting the next vehicle, the steering angle neutral point stored immediately before is read, and the steering angle is calculated on the basis of the read steering angle neutral point.

  For example, in a vehicle parked in a cold region, the yaw rate sensor is cold. Under these circumstances, when the ignition switch is turned on, the temperature of the yaw rate sensor rises due to the operation of the air conditioner or the like. However, if the ignition switch is turned off before the temperature of the yaw rate sensor is stabilized and the learned value of the steering angle neutral point is stored during that time, the learned steering angle neutral point will deviate from the correct value. . As a result, an incorrect steering angle neutral point is read when the next ignition switch is turned on.

  When steering assist control is performed, for example, a feedforward control term set from the shape of the target travel line and a deviation of the actually controlled state value from the target value for causing the host vehicle to travel along the target travel line The target rudder angle may be calculated by an arithmetic expression including a feedback control term set in accordance with the value and an integral control term set by integrating the value of the feedback control term. In this case, the feedback control term works so as to make the deviation amount that the vehicle could not follow the target travel line in the feedforward control term zero, and the integral control term could not be absorbed by the feedback control term. It works to absorb steady-state deviations (steady-state deviations generated by detection errors of various sensors).

  The integral control term is used to gradually absorb the steady-state deviation over time. For this reason, the control gain of the integral control term is set to a small value. Therefore, immediately after the start of the steering assist control, the integral control term does not function effectively.

  For this reason, when the learned steering angle neutral point is deviated, the vehicle may deviate from the lane immediately after the start of the steering assist control.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to suppress the vehicle from deviating from the lane due to the deviation of the steering angle neutral point.

In order to achieve the above object, the features of the present invention are:
Neutral point learning means (50) for learning the steering angle neutral point (θ0) of the steering mechanism using the detected value of the yaw rate sensor (60) whose detection characteristics change according to temperature;
Rudder angle detecting means (71) for obtaining a rudder angle neutral point learned by the neutral point learning means, and detecting a rudder angle (θh) based on the obtained rudder angle neutral point;
The target steering angle (θ *) for the host vehicle to travel along the target travel line (Ld) is calculated, and the steering angle (θh) follows the target steering angle (θ *). In a steering assist device comprising steering assist control means (10, 72, 73) for performing steering assist control, which is control for applying steering torque to the mechanism,
The steering assist control means includes
A feedforward control term (θFF) that is set according to the shape of the target travel line and a lateral deviation feedback control component that is set according to a lateral deviation that is a deviation amount in the road width direction of the host vehicle with respect to the target travel line Integrates the lateral deviation and the feedback control term (θFB) set according to the deviation of the actually controlled state value from the target value for driving the vehicle along the target travel line including (θFBd) A target rudder angle calculating means (21, 22, 23b, 24) for calculating the target rudder angle by an arithmetic expression including an integral control term (θI) set as
In the predetermined specific period immediately after the start of the steering assist control in which the own vehicle may deviate from the lane due to the learned deviation of the steering angle neutral point, the integral control is performed compared to after the specific period has elapsed. And a control gain variable means (23a, S11 to S14) for setting the control gain of the term to a high value.

  In the steering assist device of the present invention, the neutral point learning means learns the steering angle neutral point of the steering mechanism using the detected value of the yaw rate sensor whose detection characteristic changes according to the temperature. In this case, the neutral point learning means may learn the steering angle neutral point of the steering mechanism using at least the detection value of the yaw rate sensor. For example, the neutral point learning means sets the steering angle of the steering mechanism to the steering angle neutral point in a situation where it can be estimated that the host vehicle is traveling straight, based on the detection value of the yaw rate sensor.

  The rudder angle detection means acquires the rudder angle neutral point learned by the neutral point learning means, and detects the rudder angle based on the obtained rudder angle neutral point. For example, the rudder angle detection means acquires the rudder angle neutral point learned by the neutral point learning means when the vehicle is started. The steering assist control means calculates the target rudder angle for the host vehicle to travel along the target travel line, and applies steering torque to the steering mechanism so that the rudder angle follows the target rudder angle. Implement support control.

  The detection characteristic of the yaw rate sensor changes according to its own temperature. For this reason, the steering angle neutral point learned during the period when the temperature of the yaw rate sensor is not stable may deviate from the true value.

  Therefore, the steering assist control means includes target steering angle calculation means and control gain variable means.

  The target rudder angle calculation means is set according to a feedforward control term set according to the shape (for example, curvature) of the target travel line and a lateral deviation which is a deviation amount of the host vehicle with respect to the target travel line in the road width direction. Integrate the lateral deviation with the feedback control term set according to the deviation of the actually controlled state value from the target value for driving the vehicle along the target travel line. The target rudder angle is calculated by an arithmetic expression including an integral control term set in the above.

  The feedback control term works so as to make the deviation amount that the vehicle could not follow the target travel line in the feedforward control term zero, and the integral control term is a steady deviation (which could not be absorbed in the feedback control term) It works so as to absorb the steady deviation generated by detection errors of various sensors.

  The integral control term absorbs steady deviation over time to prevent the vehicle behavior from becoming unstable when the position of the vehicle changes due to sudden disturbance such as crosswind. Used to do. For this reason, it is desirable to set the control gain in the integral control term to be small. However, when the control gain is set in such a manner, the integral control term does not function effectively immediately after the start of the steering assist control. That is, immediately after the start of the steering assist control, the integral control term is not sufficiently large, so that the steady deviation cannot be absorbed. For this reason, when the steering angle neutral point learned by the neutral point learning means is deviated, the vehicle may deviate from the lane immediately after the start of the steering assist control.

  Therefore, the control gain varying means is configured such that the vehicle is likely to deviate from the lane due to the learned deviation of the steering angle neutral point. Set the control gain of the integral control term to a high value.

  Therefore, immediately after the start of the steering assist control, the steady deviation can be absorbed quickly by using the integral control term in which a high control gain is set. Thereby, it can suppress that a vehicle deviates from a lane due to the shift | offset | difference of a steering angle neutral point.

  In the above description, in order to assist the understanding of the invention, the reference numerals used in the embodiments are attached to the constituent elements of the invention corresponding to the embodiments in parentheses. It is not limited to the embodiment defined by.

1 is a schematic system configuration diagram of a steering assist device for a vehicle according to an embodiment. 4 is a plan view showing left and right white lines LL and LR, a target travel line Ld, and a curve radius R. FIG. It is a top view showing lateral deviation Dy which is lane information at the time of carrying out lane maintenance support control, and yaw angle thetay. It is a graph showing the temperature characteristic of a yaw rate sensor. It is a graph showing the relationship between the error of a yaw rate sensor, and the shift | offset | difference of a steering angle neutral point. It is a flowchart showing a gain setting routine.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a schematic system configuration diagram of a vehicle steering assist device according to an embodiment.

  The steering assist device 1 includes a driving assist ECU 10, a neutral point learning ECU 50, and an electric power steering ECU 70. Hereinafter, the electric power steering ECU 70 is referred to as an EPS / ECU 70 (Electric Power Steering ECU). ECU is an abbreviation for Electric Control Unit. Each ECU 10, 50, 70 has a microcomputer as a main part. In the following description, “vehicle” represents a vehicle (own vehicle) on which the steering assist device 1 is mounted.

  The neutral point learning ECU 50 is a VSC • ECU that performs vehicle stability control (VSC) for ensuring the stability of the turning direction of the vehicle, and the determination result of the straight traveling state of the vehicle estimated when the VSC is executed. It has a function to learn the rudder angle neutral point using. In the present embodiment, the VSC is not a feature, and it is not necessary to explain the VSC in order to understand the present invention. Therefore, the VSC / ECU is referred to as a neutral point learning ECU 50. Only that will be described.

  The steering angle neutral point means the steering angle of the steering mechanism when the vehicle goes straight. The neutral point learning ECU 50 is connected to the yaw rate sensor 60, the vehicle speed sensor 61, the lateral acceleration sensor 62, and the steering sensor 66. The yaw rate sensor 60 outputs a yaw rate detection signal representing the yaw rate γ around the center of gravity axis generated when the vehicle turns. The vehicle speed sensor 61 outputs a vehicle speed detection signal that represents the travel speed of the vehicle (referred to as the vehicle speed V). Since the vehicle speed V is detected based on the wheel speed, a wheel speed sensor can be used instead of the vehicle speed sensor 61. When VSC is performed, a wheel speed detection signal of a four-wheel wheel speed sensor is used to detect the side slip state of the vehicle. The lateral acceleration sensor 62 outputs a lateral acceleration detection signal representing a lateral acceleration Gy that is an acceleration in the lateral direction (vehicle width direction) of the vehicle. The steering sensor 66 outputs a steering angle detection signal indicating a steering angle θsh that is the steering angle of the steering wheel.

  The rotation angle sensor 63 is incorporated in the steering mechanism and outputs a rotation angle detection signal. Although not shown, the steering mechanism represents a mechanism (for example, a steering handle, a steering shaft, a rack and pinion mechanism, a rack bar, a tie rod, and a knuckle arm) for steering the steered wheels. The rotation angle sensor 63 of the present embodiment detects, for example, a rotation angle θsp of a shaft (a pinion shaft closest to the wheel in the steering shaft) to which a pinion gear meshed with a rack bar is fixed. The detection of the rotation angle is not necessarily the rotation angle of the pinion shaft, but the detection of the rotation angle near the wheel is less affected by the dynamics (friction, etc.) of the steering mechanism and follows the target rudder angle. Therefore, in the present embodiment, the rotation angle of the pinion shaft is detected.

  The rotation angle sensor 63 outputs a rotation angle detection signal representing a relative steering angle with the rotation position when the ignition switch is turned on being 0 deg.

  The neutral point learning ECU 50 has a vehicle speed V equal to or higher than a learning permission vehicle speed Vref that is a preset vehicle speed, an absolute value | γ | of the yaw rate γ is equal to or less than a preset straight-running determination threshold value γref, and It is estimated that the vehicle is traveling straight when the straight travel determination condition that the condition that the absolute value | Gy | of the lateral acceleration Gy is equal to or less than a preset straight travel determination threshold Gyref continues for a predetermined time or longer is satisfied. To do. The neutral point learning ECU 50 stores and updates the steering wheel steering angle θsh detected by the steering sensor 66 at that time as a steering angle neutral point in the nonvolatile memory each time when the straight traveling determination condition is satisfied (learns). ). Note that the neutral point learning ECU 50 may learn the steering angle neutral point only when the steering wheel steering angle θsh is equal to or greater than the threshold when the straight traveling determination condition is satisfied. The straight travel determination condition only needs to include at least a condition that the absolute value | γ | of the yaw rate γ is equal to or less than a preset straight travel determination threshold γref, and other additional conditions are optional. Can be set to

  The neutral point learning ECU 50 performs the above-described learning of the steering angle neutral point during the period when the ignition switch is on. When the ignition switch is turned off and turned on again, the neutral point learning ECU 50 reads the learning value stored in the nonvolatile memory and supplies the EPS / ECU 70 with information representing the steering angle neutral point θ0 learned immediately before. Each time the neutral point learning ECU learns the rudder angle neutral point θ0, the neutral point learning ECU supplies information representing the learned rudder angle neutral point θ0 to the EPS • ECU 70.

  The detected value of the yaw rate sensor 60 changes according to its own temperature. Therefore, the neutral point learning ECU 50 is, for example, a zero point of the output of the yaw rate sensor 60 when it is estimated that the yaw rate is not generated in the vehicle, for example, when the vehicle is stopped (vehicle speed V = 0). Correct the deviation. That is, a zero point correction process is performed on the output signal of the yaw rate sensor 60 in a situation where it is estimated that no yaw rate is generated in the vehicle, assuming that the yaw rate γ is zero.

The EPS / ECU 70 is a control device for an electric power steering device, and includes a steering wheel steering angle calculation unit 71 and a motor control unit 72. The steering wheel steering angle calculation unit 71 acquires the steering wheel neutral point θ0 supplied from the neutral point learning ECU 50 when the ignition switch is turned on. The steering wheel steering angle calculation unit 71 includes a steering wheel steering angle θsh detected by the steering sensor 66, a steering angle neutral point θ0 supplied from the neutral point learning ECU 50, and a rotation angle θsp of the pinion shaft detected by the rotation angle sensor 63. (Referred to as the pinion angle θsp) is used to calculate the corrected steering angle. The corrected steering angle is calculated by the following equation.
Corrected steering angle = θsp− (θsh−θ0)

  The calculation of the corrected steering angle is performed only once immediately after the ignition switch is turned on. That is, thereafter, the steering angle neutral point θ0 supplied from the neutral point learning ECU 50 to the EPS / ECU 70 is not used for calculating the corrected steering angle. The steering wheel steering angle calculation unit 71 calculates a steering wheel conversion steering angle θh obtained by converting the pinion angle sp into the rotation angle of the steering wheel. The steering wheel conversion steering angle θh is calculated by subtracting the corrected steering angle from the pinion angle θsp (θh = θsp−corrected steering angle). The steering wheel conversion steering angle θh is an absolute angle with reference to the corrected steering angle. In other words, the steering wheel conversion steering angle θh represents a steering wheel steering angle (absolute angle) detected using the learned steering angle neutral point θ0 as a reference (steering angle zero). The steering direction of the steering wheel conversion steering angle θh is specified by its sign (positive or negative).

  In the present embodiment, since the target steering angle in the LTA control described later is represented by the rotation angle of the steering handle, the rotation angle of the shaft is converted into the steering wheel conversion steering angle θh. Is expressed by the rotation angle of the other part, the steering angle using the rotation angle of that part may be calculated.

  The steering wheel steering angle calculator 71 uses the steering wheel neutral point θ0 acquired from the neutral point learning ECU 50 when the ignition switch is turned on to calculate the steering wheel conversion steering angle θh during the period until the ignition switch is turned off.

  The motor control unit 72 is connected to a torque sensor 64 provided on the steering shaft. The torque sensor 64 outputs a detection signal representing the steering torque Tr input to the steering wheel. The left and right direction of the steering torque Tr is specified by its sign. The motor control unit 72 includes a motor drive circuit (not shown), and is connected to the assist motor 73 via the motor drive circuit. The assist motor 73 is assembled to the steering mechanism, and changes the steering angle of the steered wheels by generating a steering torque.

  The motor control unit 72 detects the steering torque input by the driver to the steering handle (not shown) by the torque sensor 64, and drives and controls the assist motor 73 based on the steering torque, whereby the steering assist torque is applied to the steering mechanism. To assist the steering operation so that the driver's steering wheel operation becomes lighter.

  Further, when receiving a steering command from the driving support ECU 10, the motor control unit 72 drives the assist motor 73 with a control amount specified by the steering command (target steering angle θ *) to generate a steering assist torque. . The steering assist torque is applied to the steering mechanism by a steering command from the driving assistance ECU 10 without requiring the driver's steering operation force, unlike the steering assist torque applied to lighten the driver's steering operation. Represents the torque to be applied.

  The motor control unit 72 inputs the steering wheel conversion steering angle θh calculated by the steering wheel steering angle calculation unit 71 and the target steering angle θ * transmitted from the driving support ECU 10, and the steering wheel conversion steering angle θh is the target steering angle θ. The energization of the assist motor 73 is controlled so as to follow *, and torque is generated in the assist motor 73. As a result, the traveling direction of the vehicle can be controlled in a desired direction without the driver's steering operation.

  The driving assistance ECU 10 is connected to a vehicle speed sensor 61, a camera sensor 65, and a torque sensor 64. The camera sensor 65 captures a landscape in front of the host vehicle, analyzes image data obtained by the shooting, and recognizes (detects) a white line formed on the road in front of the host vehicle. As shown in FIG. 2, the camera sensor 65 recognizes the left white line LL and the right white line LR, and sets the lane center line that is the center position of the left and right white lines LL and LR as the target travel line Ld. Further, the camera sensor 65 calculates a curve radius R of the target travel line Ld. The target travel line Ld is not necessarily set at the center position of the left and right white lines, and may be set at a position shifted in the left-right direction by a predetermined distance from the center position. Further, the target travel line Ld estimates the position of one white line when the position of the other white line can be recognized well even if the recognition state of either the left white line LL or the right white line LR is not good. Is set by

  The camera sensor 65 calculates the position and orientation of the host vehicle in the travel lane that is divided by the left white line LL and the right white line LR. For example, as shown in FIG. 3, the camera sensor 65 detects the distance Dy in the road width direction between the reference point P of the host vehicle C and the target travel line Ld, that is, the host vehicle C is in relation to the target travel line Ld. The distance Dy displaced in the road width direction is calculated. This distance Dy is called lateral deviation Dy. The reference point P can be arbitrarily set as long as it is a position at the center in the vehicle width direction of the host vehicle C and can be specified. For example, the reference point P may be a center of gravity of the vehicle.

  Further, as shown in FIG. 3, the camera sensor 65 has an angle formed between the direction of the target travel line Ld and the direction of the host vehicle C, that is, the direction of the host vehicle C with respect to the direction of the target travel line Ld. The angle θy that the horizontal direction is shifted in the horizontal direction is calculated. This angle θy is called the yaw angle θy.

  The camera sensor 65 supplies information (R, Dy, θy) representing the calculated curve radius R, lateral deviation Dy, and yaw angle θy to the driving assistance ECU 10. The left and right directions of the curve radius R, the lateral deviation Dy, and the yaw angle θy are specified by their signs. In addition, the camera sensor 65 supplies information representing the recognition levels (recognized distances) of the left and right white lines LL and LR to the driving support ECU 10. Information supplied from the camera sensor 65 to the driving support ECU 10 is referred to as lane information.

  In the present embodiment, the camera sensor 65 calculates the lane information by calculation, but instead, the driving assistance ECU 10 may determine the lane information by calculation. In this case, the camera sensor 65 supplies image data to the driving support ECU 10, and the driving support ECU 10 obtains lane information by calculation based on the image data.

  The driving assistance ECU 10 is an electronic control device that performs lane keeping assistance control. Lane maintenance assist control is called LTA (Lane Trace Assist) control, and steering assist torque is applied to the steering mechanism so that the travel position of the host vehicle is maintained near the target travel line (for example, the center position of the travel lane). This control is given to assist the driver's steering operation. Hereinafter, the lane keeping support control is referred to as LTA control.

  LTA control is a case where LTA control is requested by an operation of a setting controller (not shown), the vehicle speed V is within the LTA-permitted vehicle speed range, and the white line recognition level satisfies the reference level, etc. The start condition is satisfied.

  The driving assistance ECU 10 includes, as a main part, a microcomputer that calculates a target rudder angle θ *, which is a target control amount for LTA control. When attention is paid to the function of the microcomputer, as shown in FIG. And a feedback control unit 22, an integration control unit 23, and an addition unit 24. In the figure, feedforward is represented as FF, and feedback is represented as FB.

As shown in the following equation (1), the target rudder angle θ * is a feedforward control term θFF that is set according to the curvature of the target travel line Ld, and a deviation amount in the road width direction of the host vehicle with respect to the target travel line Ld. A lateral deviation feedback control term θFBd set according to the lateral deviation Dy, and a yaw angle feedback control term θFBy set according to the yaw angle θy which is a deviation angle of the direction of the host vehicle with respect to the target travel line Ld, It is calculated by adding the integral control term θI obtained by integrating the lateral deviation Dy with time.
θ * = θFF + θFBd + θFBy + θI (1)

  The lateral deviation feedback control term θFBd and the yaw angle feedback control term θFBy are elements that constitute the feedback control term. Hereinafter, both may be collectively referred to as a feedback control term θFB (θFB = θFBd + θFBy).

  Note that the feedback control term θFB is not only the lateral deviation feedback control term θFBd and the yaw angle feedback control term θFBy, but also, for example, deviation of the vehicle yaw rate, deviation of the lateral velocity of the vehicle, deviation of the lateral acceleration of the vehicle, steering A feedback control term using angular deviation or the like can be arbitrarily added. In this case, detection signals from sensors such as the yaw rate sensor 60 and the lateral acceleration sensor 62 that are necessary for detecting the deviation are supplied to the driving support ECU 10, and a feedback control amount corresponding to the deviation between the target value and the sensor detection value is provided. Calculated.

  For example, the feedback control unit 22 calculates an arbitrary target value (one or more) such as a target yaw rate, a target lateral acceleration, and a target lateral velocity from a map or the like according to the value of (θFF + θFBd + θFBy + θI), A value obtained by multiplying the deviation between the target value and the sensor detection value by the feedback control gain may be added to the feedback control term θFB. Alternatively, a value obtained by multiplying the deviation between the target steering angle θ * and the steering wheel conversion steering angle θh one calculation cycle before by the feedback control gain may be added to the feedback control term θFB.

The feedforward control unit 21 calculates the feedforward control term θFF by an arithmetic expression shown in the following expression (2).
θFF = (V ² / R) × K1 (2)
Here, V is the vehicle speed, R is the radius of curvature of the target travel line, and K1 is the feedforward control gain.

The feedback control unit 22 calculates the feedback control term θFB by the calculation formula shown in the following formula (3).
θFB = θFBd + θFBy (3)
θFBd = Dy × K2 (3-1)
θFBy = θy × K3 (3-2)
Here, K2 represents a lateral deviation feedback control gain, and K3 represents a yaw angle feedback control gain. The target lateral deviation in the lateral deviation feedback control is zero, and the target yaw angle in the yaw angle feedback control is zero. Therefore, in Expressions (3-1) and (3-2), (Dy, θy) supplied from the camera sensor 65 is used as a deviation from the target value.

The integral control unit 23 computes the integral control term θI using an arithmetic expression shown in the following equation (4).
θI = θI (n−1) + Dy × t × K4 (4)
Here, θI (n−1) represents the integral control term θI one calculation cycle before, t represents the calculation cycle, and K4 represents the lateral deviation integral control gain. The lateral deviation integral control gain K4 functions as a constant that sets the degree to which the integral control term θI can be changed per unit time (per calculation cycle). The integral control term θI is a value obtained by integrating the lateral deviation Dy in proportion to the lateral deviation integral control gain K4. Accordingly, the larger the lateral deviation integral control gain K4 is, the faster the lateral deviation Dy is accumulated.

  The integral control unit 23 basically calculates the integral control term θI by the above-described calculation method. However, as will be described later, the vehicle immediately after the start of the LTA control due to the deviation of the steering angle neutral point. In order to prevent the vehicle from deviating from the lane, the function of variably setting the lateral deviation integral control gain K4, and the rate of change of the integral control term θI and the limitation of the upper limit value are defined as the lateral deviation integral control gain. A function of switching according to K4 is provided. The integration control unit 23 is roughly composed of a gain setting unit 23a, an integration calculation unit 23b, and a limiting unit 23c.

  The gain setting unit 23a sets a value of the lateral deviation integral control gain K4 in the above equation (4) by executing a gain setting routine described later, and the set lateral deviation integral control gain K4 is set as the integral calculation unit. 23b and the limiter 23c. This lateral deviation integral control gain K4 is prepared in two types, K4normal that is normally applied and K4high that is set to a value larger than K4normal, one of which is alternatively selected by the gain setting unit 23a. Is done.

  The integral calculation unit 23b calculates the integral control term θI by applying the lateral deviation integral control gain K4 (K4normal or K4high) supplied from the gain setting unit 23a to the above equation (4), and calculates the calculation result to the limiting unit. 23c. The limiting unit 23c uses a value obtained by subjecting the integral control term θI supplied from the integral calculation unit 23b to a change rate limitation and an upper limit limitation as a final calculation result of the integration control unit 23. A certain integral control term θI is output to the adder 24. The change rate limit and the upper limit limit are switched according to the set state of the lateral deviation integral control gain K4 (K4normal or K4high) supplied from the gain setting unit 23a.

  The adding unit 24 calculates the target steering angle θ * by adding the feedforward control term θFF, the feedback control term θFB (= θFBd + θFBy), and the integral control term θI, as shown in the above equation (1). .

  If the steering angle of the steering mechanism is controlled by the EPS / ECU 70 so that the steering wheel conversion steering angle θh follows the target steering angle θ * (if the assist motor 73 is controlled), the target vehicle is basically targeted. It should be possible to travel along the travel line Ld. However, if the steering angle neutral point θ0 used by the EPS • ECU 70 (the steering angle neutral point θ0 acquired from the neutral point learning ECU 50 when the ignition switch is turned on) is deviated, the steering wheel conversion steering angle θh It will shift. The deviation of the steering angle neutral point represents the deviation of the steering mechanism from the true steering angle neutral point.

  The integral control term θI represents a control amount provided to absorb a steady deviation that could not be absorbed by the feedback control term. This steady deviation occurs when detection errors (sensor errors) occur in various sensors. When the position of the vehicle body changes due to a sudden disturbance such as a crosswind, the integral control term θI is a constant deviation over time so that the vehicle behavior does not become unstable when the disturbance disappears. Used to absorb. For this reason, it is desirable that the control gain in the integral control term θI is set to a smaller value.

  When the steering angle neutral point θ0 is deviated, the steady deviation becomes large. For this reason, at the start of LTA control, the integral control term θI is not sufficiently large with respect to a large steady deviation (since the initial value is zero), and thus the steady deviation cannot be absorbed. Therefore, the vehicle may deviate from the lane.

  Here, the cause of the deviation of the rudder angle neutral point will be described.

  As described above, the detected value of the yaw rate sensor 60 changes according to its own temperature. The neutral point learning ECU 50 corrects the deviation of the zero point of the output of the yaw rate sensor 60 when the vehicle is stopped. However, as shown in FIG. 4, in a situation where the temperature Ty of the yaw rate sensor 60 is changing, the detected value γ of the yaw rate also changes in accordance with the temperature change. For example, immediately after the vehicle is started (ignition switch is turned on: IG-ON), the temperature of the yaw rate sensor 60 rises due to the start of the operation of the air conditioner, etc., so the detection characteristics of the yaw rate sensor 60 change. If the temperature of the yaw rate sensor 60 is stabilized, the zero point correction of the yaw rate sensor 60 is performed, so that the detection value of the yaw rate sensor 60 finally becomes a correct value. However, in a situation where the temperature of the yaw rate sensor 60 is changing, the detection value of the yaw rate sensor 60 is not an accurate value.

  Based on the yaw rate γ detected by the yaw rate sensor 60, the neutral point learning ECU 50 determines the steering angle steering point θsh detected by the steering sensor 66 when the vehicle is estimated to travel straight ahead as the steering angle neutral point θ0. As non-volatile memory. For this reason, if the detection value of the yaw rate sensor 60 is not accurate, an error occurs in the estimation of the straight traveling of the vehicle performed by the neutral point learning ECU 50, and as a result, the steering angle neutral point θ0 is changed from the true steering angle neutral point. It will shift. As shown in FIG. 5, the deviation of the steering angle neutral point increases as the error of the yaw rate sensor 60 increases.

  For example, in a vehicle parked in a cold region, the yaw rate sensor 60 is cold. Under these circumstances, when the ignition switch is turned on, the temperature of the yaw rate sensor 60 rises due to the operation of an air conditioner or the like. However, when the ignition switch is turned off before the temperature of the yaw rate sensor 60 is stabilized and the steering angle neutral point is learned and stored during that time, the stored steering angle neutral point deviates from the correct value. End up.

  For this reason, when the next ignition switch is turned on, the EPS • ECU 70 obtains the shifted steering angle neutral point θ0 from the neutral point learning ECU 50. The EPS / ECU 70 calculates the steering wheel conversion steering angle θh by using the steering angle neutral point θ0 acquired from the neutral point learning ECU 50 when the ignition switch is turned on until the ignition switch is turned off. For this reason, the steering angle of the steering mechanism is controlled based on the steering wheel conversion steering angle θh with reference to the shifted steering angle neutral point θ0 until the ignition switch is turned off.

  When the steering angle neutral point θ0 is deviated, the value of the integral control term θI that absorbs the steady-state deviation is small immediately after the start of the LTA control, so that the vehicle may deviate from the lane.

  Therefore, immediately after the start of the LTA control, the integral control unit 23 increases the value of the integral control term θI quickly, thereby effectively using the integral control term and suppressing the vehicle from deviating from the lane. Hereinafter, a process performed by the integration control unit 23 will be described.

  FIG. 6 shows a gain setting routine executed by the integration control unit 23. The integration control unit 23 starts the gain setting routine simultaneously with starting the LTA control.

  When the gain setting routine is started, the integral control unit 23 determines whether or not a condition for starting to set the lateral deviation integral control gain K4 to K4high (referred to as an integral high start condition) is satisfied in step S11.

The integration control unit 23 determines that the integration high start condition is satisfied when all of the following five conditions (1-1 to 1-5) are satisfied.
Condition 1-1: The vehicle speed V is higher than the threshold value Vref (V> Vref).
Condition 1-2: The lane change support control is not in progress.
Condition 1-3: The white line recognition level is equal to or greater than a threshold value.
Condition 1-4: The steering torque Tr is not more than the threshold value Trref (Tr ≦ Trref).
Condition 1-5: The integration high end condition is not satisfied.

  When the vehicle speed is low, the integral control term θI is increased to the maximum value without departing from the lane, so there is no need to increase the integral control term θI quickly. Therefore, when the vehicle speed is equal to or lower than the threshold value Vref, it is excluded from the integral high start condition by the condition 1-1.

  In this embodiment, the LTA control includes lane change support control. The lane change support control is steering control for changing the lane of the host vehicle toward an adjacent lane in a direction (left-right direction) specified by the support request operation when a driver's lane change support request operation is detected. When the lane change support control is performed, since the value of the lateral deviation Dy increases, it is excluded from the integration high start condition by the condition 1-2.

  Further, when the recognition level of the white line recognized by the camera sensor 65 is low, the lateral deviation Dy may not be detected correctly, and is excluded from the integration high start condition by the condition 1-3.

  Even during LTA control, there is a possibility that the driver may hold the steering wheel and drive the vehicle toward the left or right side of the lane. In this case, since the vehicle travels according to the driver's intention, the integral control term θI does not need to be increased quickly, and is excluded from the integral high start condition by Condition 1-4.

  Further, the integration high end condition of condition 1-5 represents the integration high end condition of step S12 described later.

  If the integration high start condition is satisfied, the integration control unit 23 determines whether or not the integration high end condition is satisfied in step S12.

The integration control unit 23 determines that the integration high end condition is satisfied when any (at least one) of the following three conditions (2-1 to 2-3) is satisfied.
Condition 2-1: A predetermined time has elapsed since the integration high.
Condition 2-2: A predetermined time has elapsed since the yaw angle θy became equal to or less than the threshold value, and a predetermined time elapsed after the lateral deviation Dy became equal to or less than the threshold value.
Condition 2-3: The value of the integral control term θI has reached the upper limit value.

  Condition 2-1 means that the established condition has continued for a predetermined time or longer after the integration high start condition is established. Further, the condition 2-2 means that the duration when the yaw angle θy is equal to or less than the threshold is equal to or longer than the predetermined time, and the duration when the lateral deviation Dy is equal to or smaller than the threshold is equal to or longer than the predetermined time. doing.

  When the integration high end condition is not satisfied (S12: No), that is, when the integration high start condition is satisfied and the integration high end condition is not satisfied, the integration control unit 23 determines in step S13. The lateral deviation integral control gain K4 is set to K4high.

  On the other hand, when the integration high start condition is not satisfied (S11: No), or even when the integration high start condition is satisfied, the integration high end condition is satisfied (S11: Yes, S12: No). In step S14, the integral control unit 23 sets the lateral deviation integral control gain K4 to K4normal.

  K4high is a lateral deviation integral control gain set to increase the integral control term θI quickly only immediately after the start of the LTA control, assuming that the steering angle neutral point is deviated, and is larger than K4normal. Is set to K4normal is set to a small value so that the value of the integral control term θI does not become larger than necessary due to sudden disturbance. The lateral deviation integration control gain K4 is set to K4high only during a predetermined period (period until the integration high end condition is satisfied) when the integration high start condition is satisfied, and thereafter is set to K4normal. .

  The processes in steps S11 to S14 are performed by the gain setting unit 23a.

  After setting the lateral deviation integral control gain K4 in step S13 or step S14, the integral control unit 23 calculates the integral control term θI using the above equation (3) in step S15. The process of step S15 is a process performed by the integral calculation unit 23b. Hereinafter, an operation mode in which K4high is set as the lateral deviation integral control gain K4 is referred to as an integral high mode, and an operation mode in which K4normal is set as the lateral deviation integral control gain K4 is referred to as an integral normal mode. The period in which the integral high mode is set as the calculation mode corresponds to the specific period of the present invention.

  Subsequently, in step S <b> 16, the integral control unit 23 places a change rate limit on the integral control term θI. This change rate restriction sets an upper limit of the change rate of the integral control term θI and changes the integral control term θI within the upper limit. In the present embodiment, the magnitude of the difference between the integral control term θI (n−1) calculated one calculation cycle before and the integral control term θI (n) calculated this time (| θI (n) −θI ( n-1) |), that is, the change amount of the integral control term θI per unit time is used to set the upper limit value of this change amount. Accordingly, when the rate of change of the integral control term θI calculated in step S15 exceeds the upper limit value, the integral control term θI changed using the upper limit value is set as the integral control term θI after the rate of change limitation. If the change rate of the integral control term θI does not exceed the upper limit value, the integral control term θI calculated in step S15 is set as it is as the integral control term θI after the change rate is limited.

  The change rate limit varies depending on the calculation mode. The change rate limitation is relaxed in the integral high mode compared to the integral normal mode. That is, the upper limit value of the change rate in the integral high mode is set to a value higher than the upper limit value of the change rate in the integral normal mode. Therefore, in the integral high mode, the increase in the integral control term θI is not hindered by the change rate limitation, and the integral control term θI can be increased earlier than in the integral normal mode.

  Subsequently, in step S17, the integral control unit 23 places an upper limit on the integral control term θI. Since the integral control term θI is set with a sign (positive or negative) according to the direction, it is correctly limited to the upper limit value for the absolute value of the integral control term θI, in other words, above the integral control term θI. It is a lower limit. In this specification, when discussing the magnitude of the control amount, the absolute value is used. When the integral control term θI calculated in step S16 (θI after the change rate limit) exceeds the upper limit value, the integral control unit 23 sets the upper limit value as the integral control term θI after the upper limit limit. If the integral control term θI (θI after the rate of change restriction) does not exceed the upper limit value, the integral control term θI calculated in step S16 is set as it is as the integral control term θI after the upper limit restriction. This upper limit value is also the upper limit value in the integral high end condition 2-3.

  The upper limit value of the integral control term θI varies depending on the calculation mode. The upper limit value of the integral control term θI in the integral high mode is set to a value smaller than the upper limit value of the change rate in the integral normal mode. This is to prevent erroneous learning of the integral control term θI due to a sudden disturbance in the integral high mode (making the integral control term θI an excessive value by considering a sudden error as a steady error). .

  Subsequently, in step S18, the integral control unit 23 outputs the value of the integral control term θI calculated in step S17 to the adding unit 24. The processing in steps S16 to S18 is processing performed by the restriction unit 23c.

  Subsequently, in step S19, the integral control unit 23 determines whether or not the LTA control is finished. If the LTA control is not finished (S19: No), the processing is returned to step S11 and described above. Repeat the process. The LTA control ends when a preset LTA control end condition is satisfied.

  When the LTA control ends, the integration control unit 23 ends the gain setting routine.

  According to the steering assist device of the present embodiment described above, the establishment of the integral high start condition is determined as the LTA control starts. This integral high start condition is a situation where the vehicle is controlled by a driver's requested operation (handle operation or lane change support request), or there is a possibility that the white line recognition level is low and the LTA control cannot be performed with a predetermined accuracy. It is established with the start of LTA control except for the situation.

  Even when the steering angle neutral point is deviated, the lateral deviation can be absorbed by the integral control term θI. However, since the value of the integral control term θI is zero at the start of the LTA control, when the lateral deviation integral control gain K4 is small, the integral control term θI increases to a level that can absorb the lateral deviation. It takes time, during which the vehicle may deviate from the lane.

  Therefore, in the present embodiment, in a situation where the vehicle should be prevented from departing from the lane, the calculation mode is set to the integral high mode when the LTA control is started. As a result, the value of the lateral deviation integral control gain K4 is set to K4high, and the increasing speed of the integral control term θI is increased. Therefore, it can suppress that a vehicle deviates from a lane.

  When the integral high end condition is satisfied, the calculation mode is set to the integral normal mode. Once the integration high end condition is satisfied, the integration high start condition is not satisfied and the integration normal mode is maintained. When the integral high end condition is satisfied, the vehicle is traveling near the target travel line, or the value of the integral control term θI is sufficiently large.

  As described above, the value of the lateral deviation integral control gain K4 is set to K4high only immediately after the start of the LTA control in which the vehicle may deviate from the lane due to the deviation of the steering angle neutral point. Therefore, after the integration high termination condition is satisfied, the value of the lateral deviation integral control gain K4 is set to K4normal, so that the vehicle behavior is prevented from becoming unstable due to the integral control term θI due to sudden disturbance. Can do.

  As mentioned above, although the steering assistance apparatus of the vehicle which concerns on this embodiment was demonstrated, this invention is not limited to the said embodiment, A various change is possible unless it deviates from the objective of this invention.

  For example, in the present embodiment, a white line is detected by the camera sensor 65, and the target travel line Ld is set based on the white line. For example, when a preceding vehicle exists in front of the host vehicle, the preceding vehicle It is also possible to set the travel locus to the target travel line Ld. In this case, for example, the traveling locus of the preceding vehicle can be acquired by detecting the preceding vehicle using a millimeter wave radar, a rider (LIDAR), or the like.

  Further, in the present embodiment, the EPS / ECU 70 uses the steering angle neutral point acquired from the neutral point learning ECU 50 when the ignition switch is turned on until the ignition switch is turned off, but the ignition switch is turned on. In the middle of the period, the neutral point learning ECU 50 may update the latest steering angle neutral point learned.

  In this embodiment, the integration control unit 23 calculates the integral of the lateral deviation Dy, but calculates the value obtained by adding the integral of the other deviation such as the yaw angle θy to the integral of the lateral deviation Dy. Also good.

  DESCRIPTION OF SYMBOLS 1 ... Steering assistance apparatus, 10 ... Driving assistance ECU, 21 ... Feedforward control part, 22 ... Feedback control part, 23 ... Integration control part, 23a ... Gain setting part, 23b ... Integration calculating part, 23c ... Limiting part, 24 ... Addition unit, 50 ... neutral point learning ECU, 60 ... yaw rate sensor, 61 ... vehicle speed sensor, 62 ... lateral acceleration sensor, 63 ... rotation angle sensor, 64 ... torque sensor, 65 ... camera sensor, 66 ... steering sensor, 70 ... EPS ECU, 71: Steering angle calculation unit, 72: Motor control unit, 73: Assist motor, R ... Curve radius, Dy ... Lateral deviation, θy ... Yaw angle, γ ... Yaw rate, V ... Vehicle speed, θsp ... Rotation angle, θsh: Steering wheel steering angle, Tr: Steering torque, Ld: Target travel line, LL, LR ... White line, θ *: Target steering angle, θh: Steering wheel steering steering angle, θ0: Middle steering angle Point, [theta] ff ... feedforward control term, .theta.fb ... feedback control term, θFBd ... lateral deviation feedback control term, θFBy ... yaw angle feedback control term, .theta.I ... integral control term, K4, K4high, K4normal ... lateral deviation integral control gain.

Claims (1)

Neutral point learning means for learning the steering angle neutral point of the steering mechanism using the detected value of the yaw rate sensor whose detection characteristic changes according to temperature;
Rudder angle detecting means for obtaining a rudder angle neutral point learned by the neutral point learning means, and detecting a rudder angle based on the obtained rudder angle neutral point; and
Steering support control, which is a control for calculating a target rudder angle for the host vehicle to travel along the target travel line and applying steering torque to the steering mechanism so that the rudder angle follows the target rudder angle. In a steering support device comprising:
The steering assist control means includes
A feedforward control term set according to the shape of the target travel line, and a lateral deviation feedback control component set according to a lateral deviation which is a deviation amount of the vehicle in the road width direction with respect to the target travel line. A feedback control term set in accordance with a deviation of an actually controlled state value from a target value for causing the vehicle to travel along the target travel line; and an integral control term set by integrating the lateral deviation; , A target rudder angle calculating means for calculating the target rudder angle by an arithmetic expression including
In the predetermined specific period immediately after the start of the steering assist control in which the own vehicle may deviate from the lane due to the learned deviation of the steering angle neutral point, the integral control is performed compared to after the specific period has elapsed. A steering assist device comprising: control gain variable means for setting the control gain of the term to a high value.

Images