Control device and control method for a vehicle

Abstract

The present disclosure relates to a control device and a control method for a vehicle. The control device for a vehicle is configured to execute a first control that outputs a steering torque based on a predetermined first gain and a second control that outputs a steering torque based on a predetermined second gain different from the first gain. In the execution of the first control, a start prediction time that is a prediction time until a start condition for the second control is satisfied is calculated, and in the case where the start prediction time satisfies a predetermined condition, a reduction control that causes the output steering torque of the first control to approach 0 is executed before the start of the second control.

Description

Technical Field

[0001] This disclosure relates to vehicle control devices and control methods. Background Technology

[0002] Japanese Patent Application Publication No. 2019-209844 discloses a driving assistance device that predicts the amount of driver's operation in response to the external environment, calculates the deviation between the predicted amount and the driver's actual amount of operation, and optimizes the amount of driver's operation by changing the reaction force characteristics of the operating part to the reaction force characteristics corresponding to the deviation. Summary of the Invention

[0003] In driver assistance systems capable of executing multiple steering controls as driving aids, there are situations where the currently executing steering control is switched to another steering control. Generally, multiple steering controls have different purposes, and therefore different rudder angle following characteristics relative to the target rudder angle. Therefore, when switching from the currently executing steering control to another steering control with different rudder angle following characteristics, the following problem arises: with the switch, the steering torque changes abruptly, causing discomfort (a sense of unease) for vehicle occupants, including the driver.

[0004] One of the purposes of this disclosure is to provide a technique that can effectively suppress changes in steering torque as control switches change.

[0005] This disclosure relates to a vehicle control device and a control method configured to perform a first control based on a predetermined first gain output steering torque and a second control based on a predetermined second gain output steering torque different from the first gain. During the execution of the first control, a start prediction time is calculated as a prediction time up to the fulfillment of the start condition of the second control, and if the start prediction time meets the predetermined condition, a retraction control is performed to bring the output steering torque of the first control close to 0 before the start of the second control.

[0006] Based on the control device and method described above, during the execution of the first control, when switching the steering control to a second control with a gain different from the first control, a retraction control is executed before starting the second control to bring the output steering torque of the first control close to zero. This effectively suppresses fluctuations in steering torque due to control switching and also effectively prevents discomfort to the vehicle's occupants. Attached Figure Description

[0007] Hereinafter, the features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described with reference to the accompanying drawings, in which the same reference numerals denote the same elements, and wherein:

[0008] Figure 1 This is a schematic diagram illustrating the hardware configuration of the vehicle involved in this embodiment;

[0009] Figure 2 This is a schematic diagram illustrating the software configuration of the control device according to this embodiment;

[0010] Figure 3 This is a diagram illustrating driving lane information;

[0011] Figure 4 This is a timing chart illustrating the switching control involved in this embodiment;

[0012] Figure 5 This is a flowchart illustrating the switching control routine involved in this embodiment;

[0013] Figure 6 This is a timing diagram illustrating the switching control of the comparative example. Detailed Implementation

[0014] Hereinafter, with reference to the accompanying drawings, the vehicle control device and control method according to this embodiment will be described.

[0015] Hardware configuration

[0016] Figure 1 This is a schematic diagram illustrating the hardware configuration of a vehicle SV to which the control device involved in this embodiment is applied.

[0017] The vehicle SV has an Electronic Control Unit (ECU) 10. The ECU 10 includes a Central Processing Unit (CPU) 11, Read Only Memory (ROM) 12, Random Access Memory (RAM) 13, and an Interface Device (I / F) 14. The CPU 11 is a processor that executes various programs stored in the ROM 12. The ROM 12 is non-volatile memory that stores data required by the CPU 11 to execute various programs. The RAM 13 is volatile memory that provides the working area for various programs to be expanded when executed by the CPU 11. The Interface Device 14 is a communication device for communicating with external devices.

[0018] ECU10 is a central device for driving assistance, serving as the driver's control center for assisted driving operations in the vehicle SV. As a driving assistance system, ECU10 can execute a normal assistance mode with a low degree of steering intervention and an avoidance assistance mode with a high degree of steering intervention, and can switch between the normal assistance mode and the avoidance assistance mode.

[0019] The general assistance mode refers to the mode that provides driving assistance to the driver. In this embodiment, reaction force control is performed to modify the reaction force characteristics of the steering mechanism so that the driver's steering input is an appropriate input. The avoidance assistance mode refers to the mode that provides driving assistance to the main body of the system. In this embodiment, lane departure warning control (hereinafter, LDA control) is performed to prevent the vehicle from deviating from its driving lane. Details about these reaction force controls and LDA control will be explained later.

[0020] The ECU10 is communicatively connected to a drive unit 20, a braking unit 21, a steering unit 22, an internal sensor unit 30, an external sensor unit 40, etc.

[0021] The drive unit 20 generates driving force that is transmitted to the drive wheels of the vehicle SV. Examples of drive units 20 include electric motors and engines. In this embodiment, the vehicle SV can be any of a hybrid electric vehicle (HEV), a plug-in hybrid electric vehicle (PHEV), a fuel cell electric vehicle (FCEV), a battery electric vehicle (BEV), and a motor-driven vehicle. The braking unit 21 applies braking force to the wheels of the vehicle SV.

[0022] The steering mechanism 22 applies a steering force to the wheels of the vehicle SV. The steering mechanism 22 can be either a rack and pinion type or a steer-by-wire type. The steering mechanism 22 includes a steering operation unit 23, including a steering wheel SW. Additionally, the steering mechanism 22 includes a steering motor 25 that applies steering torque to the steering shaft 24. The steering motor 25 generates steering torque according to commands from the ECU 10. This steering torque enables the left and right steering wheels of the vehicle SV to turn, thereby applying a reaction force to the steering wheel SW relative to the driver's operation. Furthermore, the steering operation unit 23 is not limited to the steering wheel SW and can be a shape other than a steering stick or a wheel.

[0023] The interior boundary sensor device 30 is a type of sensor used to detect the state of the vehicle's spatial dimension (SV). Specifically, the interior boundary sensor device 30 includes a vehicle speed sensor 31, a steering angle sensor 32, a yaw rate sensor 33, and an acceleration sensor 34.

[0024] Vehicle speed sensor 31 detects the vehicle speed (V) of vehicle SV. Steering angle sensor 32 detects the rotation angle of the steering wheel or steering shaft (not shown), i.e., the steering angle, of vehicle SV. Yaw rate sensor 33 detects the yaw rate of vehicle SV. Acceleration sensor 34 detects the acceleration of vehicle SV. The interior sensor device 30 sends the status of vehicle SV detected by each sensor 31 to 34 to ECU 10 at predetermined intervals.

[0025] The external sensor device 40 is a type of sensor that identifies target information related to objects around the vehicle SV. Specifically, the external sensor device 40 includes a radar sensor 41, a camera sensor 42, etc. Here, target information may include, for example, surrounding vehicles, road markings such as white lines painted on the road, curbs, guardrails, walls, etc. The external sensor device 40 repeatedly sends the acquired target information to the ECU 10 at predetermined intervals.

[0026] Radar sensor 41 is disposed, for example, at the front of vehicle SV, to detect targets present in the area in front of vehicle SV. Radar sensor 41 includes millimeter-wave radar and / or lidar. Millimeter-wave radar emits millimeter-wave radio waves (millimeter waves) and receives millimeter waves (reflected waves) reflected by targets present within the emission range. Based on the phase difference between the emitted millimeter waves and the received reflected waves, the attenuation level of the reflected waves, and the time from the emission of the millimeter waves to the reception of the reflected waves, millimeter-wave radar obtains the relative distance and relative speed between vehicle SV and targets. Lidar obtains the shape of targets detected in front of vehicle SV, the relative distance and relative speed between vehicle SV and targets by sequentially scanning pulsed laser light with wavelengths shorter than millimeter waves in multiple directions and receiving reflected light reflected by targets.

[0027] The camera sensor 42 is, for example, a stereo camera and / or a monocular camera, and can be a digital camera with imaging elements such as CMOS or CCD. The camera sensor 42 is, for example, mounted on the upper part of the windshield of the vehicle SV. The camera sensor 42 captures images of the area in front of the vehicle SV, processes the captured image data, and thereby obtains information about objects in front of the vehicle SV. This object information includes information such as the type of object detected in front of the vehicle SV, the relative distance between the vehicle SV and the object, and the relative speed between the vehicle SV and the object. The type of object can be identified, for example, through machine learning such as pattern matching.

[0028] Software composition

[0029] Figure 2 This is a schematic diagram illustrating the software configuration of the ECU10 according to this embodiment. For example... Figure 2As shown, the ECU 10 includes lane recognition unit 100, reaction force control unit 110, LDA control unit 120, and switching control unit 130 as functional elements. These functional elements 100 to 130 are implemented by the CPU 11 of the ECU 10 reading programs stored in the ROM 12 into the RAM 13 and executing them. Furthermore, in this embodiment, each functional element 100 to 130 is described as being integrated into the ECU 10 as a single piece of hardware; however, any part of them can also be housed in a separate ECU. Additionally, all or part of each functional element 100 to 130 of the ECU 10 can also be housed in an information processing device of a facility capable of communicating with the vehicle's SV (e.g., a management center).

[0030] The lane recognition unit 100 identifies the driving lane of the vehicle SV based on the detection results of the external sensor device 40. Here, the driving lane refers not only to the driving area marked by white lines, yellow lines, or other markings painted on the road surface, but also to the driving area marked by structures such as curbs, guardrails, and walls. Furthermore, for convenience, the boundaries of the driving areas marked by these markings, structures, etc., will be referred to as "boundary lines" below.

[0031] like Figure 3 As shown, the lane recognition unit 100 recognizes the left boundary line LL and the right boundary line LR. Furthermore, the lane recognition unit 100 calculates the curve radius R of the center line LC, which is the center position of these left and right boundary lines LL and LR, and calculates the yaw angle θy between the direction of the center line LC and the direction in which the vehicle SV is facing. Moreover, the lane recognition unit 100 calculates the respective distances (hereinafter, lateral deviation Δx) in the road width direction between the left front wheel of the vehicle SV and the left boundary line LL, and between the right front wheel and the right boundary line LR. Figure 3 This only represents the lateral deviation Δx between the left front wheel and the left boundary line LL. In this case, there are two lateral deviations Δx, one on the left and one on the right. During LDA control, which will be described later, the lateral deviation Δx is used, which is assumed to be the direction in which the vehicle SV deviates from the driving lane, that is, the direction represented by the yaw angle θy. Hereinafter, the lateral deviation Δx, yaw angle θy, and curve radius R calculated by the lane recognition unit 100 will be collectively referred to as driving lane information.

[0032] When the driver assistance is in normal assistance mode, the reaction force control unit 110 performs reaction force control by changing the reaction force characteristics of the steering operation unit 23 to make the amount of steering wheel SW operated by the driver appropriate while maintaining the driver's sense of control. The reaction force characteristics refer to the characteristics of the reaction force applied to the steering shaft 24 by the steering motor 25 according to the amount of steering wheel SW operated by the driver.

[0033] The reaction force control unit 110 predicts the appropriate amount of steering input from the driver based on the detection results from the inner boundary sensor device 30 and the driving lane information obtained by the lane recognition unit 100. The appropriate amount of steering input refers to the amount of steering wheel movement (SW) that the driver would typically make in response to the conditions surrounding the vehicle (SV). The reaction force control unit 110 compares the predicted appropriate amount of steering input with the driver's actual amount of steering input (actual amount of steering input). The driver's amount of steering input can be obtained based on the detection results from the steering angle sensor 32.

[0034] When the actual operating amount is appropriate, the reaction force control unit 110 maintains the reaction force characteristic of the steering operation unit 23 at a predetermined reference reaction force characteristic. Conversely, when the actual operating amount is not appropriate, the reaction force control unit 110 changes the reaction force characteristic of the steering operation unit 23 from the reference reaction force characteristic to a subject-feeling maintenance reaction force characteristic. The subject-feeling maintenance reaction force characteristic is a reaction force characteristic that maintains the pilot's subject-feeling while making it easy for the actual operating amount to remain at an appropriate operating amount; it can be set by making a small change to the reference reaction force characteristic. For example, the subject-feeling maintenance reaction force characteristic can be set by making the reaction force change when the actual operating amount is at an appropriate operating amount larger than the reaction force change when the actual operating amount deviates from or seems to deviate from the appropriate operating amount.

[0035] If the reaction force control unit 110 changes the reaction force characteristic to a subject-sensing sustained reaction force characteristic, it calculates the target rudder angle (hereinafter, the reaction force target rudder angle σA) for reaction force control based on the subject-sensing sustained reaction force characteristic. Furthermore, the reaction force control unit 110 calculates the reaction force auxiliary torque τSA (=τA(σA-σD)) by multiplying the rudder angle difference Δσ between the reaction force target rudder angle σA and the actual rudder angle σD obtained by the steering angle sensor 32 by a predetermined torque gain (hereinafter, reaction force torque gain τA). The reaction force control unit 110 then sends a command signal containing information representing the reaction force auxiliary torque τSA to the steering device 22. As a result, the reaction force auxiliary torque τSA is transmitted from the steering motor 25 to the steering shaft 24, applying the desired reaction force to the steering wheel SW.

[0036] When the vehicle SV appears to be deviating from its driving lane, the LDA control unit 120 performs LDA control to suppress the vehicle SV's deviation from the driving lane. Based on the driving lane information (Δx, θy, R) obtained by the lane recognition unit 100, the LDA control unit 120 calculates the target steering angle for LDA control (hereinafter, LDA target steering angle σB). The LDA target steering angle σB is a steering angle set to prevent the vehicle SV from deviating to the outside of the boundary lines LL and LR.

[0037] The LDA control unit 120 determines whether an LDA start condition is satisfied. Examples of the LDA start condition include a case where a predicted arrival time TR until the vehicle SV reaches the boundary lines LL and LR becomes shorter than a predetermined threshold time Tv (TR<Tv). The predicted arrival time TR can be obtained based on the following equation (1) that assumes the vehicle SV makes a linear motion with constant acceleration until reaching the boundary lines LL and LR.

[0038]

[0039] In equation (1), Δx is a lateral deviation, V is the vehicle speed of the vehicle SV, and YR is the yaw rate of the vehicle SV.

[0040] When the LDA start condition is satisfied, the LDA control unit 120 calculates an LDA assist torque τSB (=τB(σB-σD)) by multiplying a steering angle difference Δσ between the LDA target steering angle σB and the actual steering angle σD acquired by the steering angle sensor 32 by a predetermined torque gain (hereinafter, referred to as LDA torque gain τB). In addition, when calculating the LDA assist torque τSB, the LDA control unit 120 transmits a command signal including information indicating the LDA assist torque τSB to the steering device 22. As a result, the LDA assist torque τSB is transmitted from the steering motor 25 to the steering shaft 24, and the steering wheel of the vehicle SV is steered, thereby suppressing the vehicle SV from departing from the traveling lane.

[0041] Here, a case is assumed where the driving assist is switched from the normal assist mode to the avoidance assist mode, that is, the steering control is shifted from the reaction force control to the LDA control. The reaction force control is driver-oriented driving assist with low steering angle followability with respect to the target steering angle, and the LDA control is system-oriented driving assist with high steering angle followability with respect to the target steering angle. In general, the smaller the torque gain, the lower the steering angle followability with respect to the target steering angle, and the larger the steering angle difference between the target steering angle and the actual steering angle. On the other hand, the larger the torque gain, the higher the steering angle followability with respect to the target steering angle, and the smaller the steering angle difference between the target steering angle and the actual steering angle. Therefore, when switching the steering control from the reaction force control with low steering angle followability to the LDA control with high steering angle followability, it is considered that the target steering angle is gradually changed to maintain the steering angle smoothly.

[0042] Figure 6 is a time chart illustrating an example of a case where the steering angle is smoothly maintained when the steering control is switched from the reaction force control to the LDA control. Figure 6 is a comparative example of the present disclosure. Figure 6The time intervals t1 to t2 represent the gradual transition period during which the target rudder angle (reference σS) slowly changes from the reaction force target rudder angle σA to the LDA target rudder angle σB in order to smoothly maintain the rudder angle. Compared with reaction force control, LDA control has a higher rudder angle following capability, therefore the LDA torque gain τB is larger than the reaction force torque gain τA (τA < τB). Therefore, even if a gradual transition period for the target rudder angle is set, before and after the control switch, the output steering torque will change dramatically from the reaction force auxiliary torque τSA (=τA(σA-σD)) to the LDA auxiliary torque τSB (=τB(σB-σD)), resulting in a sudden torque change (reference σS). Figure 6 (α) There is a problem that sudden torque changes can cause discomfort and unpleasantness to occupants, including the driver. Additionally, there are also issues during gradual torque changes. Figure 6 The problem of the LDA target rudder angle σB deviating significantly from the actual rudder angle σD, as shown by the dashed line, thus worsening the rudder angle following performance of LDA control (see reference). Figure 6 (β in the text)

[0043] Therefore, the switching control unit 130 in this embodiment suppresses sudden torque changes during control switching by performing a pre-retreat control that brings the reaction force auxiliary torque τSA close to 0 (zero) before switching from reaction force control to LDA control. Hereinafter, based on... Figure 4 The timeline shown illustrates the details of the pre-retreat control.

[0044] exist Figure 4 At time t0, the driving assistance is in normal assistance mode, and reaction force control is being executed. During the execution of reaction force control, the switching control unit 130 repeatedly calculates the rudder angle difference Δσ between the target reaction force rudder angle σA and the actual rudder angle σD at a predetermined period. Additionally, the switching control unit 130 repeatedly calculates the prediction time (hereinafter, the start prediction time TB) from the current time until the start condition for LDA control is met at a predetermined period. The start prediction time TB is the time from the current time until the aforementioned prediction arrival time TR becomes shorter than the threshold time Tv.

[0045] The switching control unit 130 calculates the prediction time (hereinafter, the retreat prediction time TD) required to bring the current reaction force assist torque τSA to approximately 0 (zero) based on a preset torque reduction gradient τgr. The torque reduction gradient τgr is the rate of change of torque (the amount of change per unit time) that gradually reduces the reaction force assist torque τSA to 0 (zero), and is set with the reference of not causing discomfort to the driver or suppressing discomfort to a minimum. The torque reduction gradient τgr can be a fixed value or a variable value corresponding to the steering angle difference Δσ.

[0046] The switching control unit 130 calculates the retraction prediction time TD (=Δσ / τgr) by dividing the rudder angle difference Δσ by the torque drop gradient τgr. The switching control unit 130 determines whether the retraction prediction time TD is greater than or equal to the start prediction time TB. If, at time t1, the retraction prediction time TD is greater than or equal to the start prediction time TB (TD≥TB), the switching control unit 130 starts pre-retraction control from time t1.

[0047] The switching control unit 130 determines whether the rudder angle difference Δσ has become 0 (zero) through the execution of the pre-retreat control. At time t2, when the rudder angle difference Δσ becomes 0 (zero) and the reaction force auxiliary torque τSA also becomes approximately 0 (zero), the switching control unit 130 ends the pre-retreat control. At the same time, the switching control unit 130 starts LDA control, that is, switches the steering control from reaction force control to LDA control. In this way, by starting LDA control when the reaction force auxiliary torque τSA is approximately 0 (zero), the occurrence of torque fluctuations accompanying the switching can be effectively suppressed even when the rudder angle following characteristics (torque gains τA, τB) of these reaction force controls and LDA controls are different. In addition, by starting LDA control from the state where the rudder angle difference Δσ is 0 (zero), the rudder angle following characteristics of LDA control can also be effectively ensured. Furthermore, as long as there is still time remaining until the reaction force auxiliary torque τSA becomes approximately 0 (zero), reaction force control continues, thereby maximizing the working time of reaction force control.

[0048] Figure 5 This is a flowchart illustrating the switching control routine performed by CPU 11 of ECU 10. This routine begins based on the execution of reaction force control.

[0049] In S100, ECU10 determines whether reaction force control is being performed. If reaction force control is being performed (yes), ECU10 proceeds to S110. On the other hand, if reaction force control is not being performed (no), ECU10 exits the current routine (return).

[0050] In S110, ECU10 calculates the rudder angle difference Δσ between the target rudder angle σA and the actual rudder angle σD, and calculates the start prediction time TB from the current moment until the start condition for LDA control is met. Next, in S120, ECU10 determines whether the retreat prediction time TD (=Δσ / τgr), obtained by dividing the rudder angle difference Δσ by the torque gradient τgr, is greater than or equal to the start prediction time TB. If the retreat prediction time TD is greater than or equal to the start prediction time TB (yes), ECU10 proceeds to the processing in S130. On the other hand, if the retreat prediction time TD is shorter than the start prediction time TB (no), ECU10 returns to the determination process in S100.

[0051] In S130, ECU10 performs pre-retreat control based on the torque gradient τgr to make the reaction force auxiliary torque τSA approximately close to 0 (zero). Next, in S140, ECU10 determines whether the rudder angle difference Δσ has become 0 (zero) through the execution of the pre-retreat control. If the rudder angle difference Δσ has not become 0 (zero) (No), ECU10 returns to the processing in S130 and continues the pre-retreat control. On the other hand, if the rudder angle difference Δσ has become 0 (zero) (Yes), ECU10 enters the processing in S150 and begins LDA control. Afterwards, ECU10 exits the current routine.

[0052] The control device and control method of the vehicle involved in this embodiment have been described above. However, this disclosure is not limited to the above embodiment. Various modifications can be made as long as they do not depart from the purpose of this invention.

[0053] For example, in the above embodiment, the switching control unit 130 is configured to perform pre-retreat control when switching the driving assistance from normal assistance mode to avoidance assistance mode, that is, when the steering control is switched from reaction force control to LDA control. However, it can also perform pre-retreat control when switching the driving assistance from avoidance assistance mode to normal assistance mode, that is, when the steering control is switched from LDA control to reaction force control. In this case, it is also possible to suppress the sudden change (sudden decrease) in torque accompanying the control switch, and effectively prevent discomfort to the driver.

[0054] In addition, the avoidance assist mode is not limited to LDA control, but can also be other steering controls with a relatively large degree of steering intervention, such as Lane Keeping Assist (LTA control) and Precipitation Control System (PCS control).

[0055] Furthermore, the application of this disclosure is not limited to steering control, but can also be applied to the control of general actuators in power transmission systems, etc., where switching between two or more controls with different gains for a single actuator is possible. As an example of such control, switching the regenerative braking force of an electric motor to a stronger braking force can be cited.

[0056] Alternatively, the vehicle SV can also be a vehicle capable of switching between automatic and manual driving. In this case, the controls disclosed herein can be executed during manual driving.

Claims

1. A vehicle control device configured to perform a first control based on a predetermined first gain output steering torque and a second control based on a predetermined second gain different from said first gain output steering torque. During the execution of the first control, a start prediction time is calculated as the prediction time until the start condition of the second control is met. If the start prediction time meets a predetermined condition, a retraction control is executed before the second control begins, making the output steering torque of the first control approach zero. The second gain is a gain greater than the first gain, and the second control is a steering control with higher rudder angle following performance relative to the target rudder angle than the first control. The first control is a reaction force control that predicts the appropriate amount of driver input to the steering mechanism in accordance with the conditions around the vehicle, sets an appropriate operating range including the appropriate amount of input, and modifies the reaction force characteristics of the steering mechanism so that the driver's input is within the appropriate operating range. The second control is lane departure suppression control, which prevents the vehicle from deviating from its driving lane. The control device is configured such that, During the execution of the first control, the difference between the target rudder angle and the actual rudder angle is repeatedly calculated at a predetermined cycle. Furthermore, this rudder angle difference is divided by a pre-set torque gradient to calculate the retraction prediction time. This retraction prediction time is the prediction time until the output steering torque of the first control becomes 0 after the execution of the retraction control. The target rudder angle is the target rudder angle of the first control. The control device is configured such that, If the predicted rollback time is shorter than the predicted start time, it is determined that the predetermined condition is not met, and the rollback control is not initiated. If the predicted retreat time is greater than the predicted start time, it is determined that the predetermined condition is met, and the retreat control is initiated.

2. A vehicle control method configured to perform a first control based on a predetermined first gain output steering torque and a second control based on a predetermined second gain different from the first gain output steering torque, comprising: During the execution of the first control, a start prediction time is calculated as the prediction time until the start condition of the second control is met. If the start prediction time meets a predetermined condition, a retraction control is executed before the second control begins, making the output steering torque of the first control approach zero. The second gain is a gain greater than the first gain, and the second control is a steering control with higher rudder angle following performance relative to the target rudder angle than the first control. The first control is a reaction force control that predicts the appropriate amount of driver input to the steering mechanism in accordance with the conditions around the vehicle, sets an appropriate operating range including the appropriate amount of input, and modifies the reaction force characteristics of the steering mechanism so that the driver's input is within the appropriate operating range. The second control is lane departure suppression control, which prevents the vehicle from deviating from its driving lane. In the control method, During the execution of the first control, the difference between the target rudder angle and the actual rudder angle is repeatedly calculated at a predetermined cycle. Furthermore, this rudder angle difference is divided by a pre-set torque gradient to calculate the retraction prediction time. This retraction prediction time is the prediction time until the output steering torque of the first control becomes 0 after the execution of the retraction control. The target rudder angle is the target rudder angle of the first control. In the control method, If the predicted rollback time is shorter than the predicted start time, it is determined that the predetermined condition is not met, and the rollback control is not initiated. If the predicted retreat time is greater than the predicted start time, it is determined that the predetermined condition is met, and the retreat control is initiated.

Images