Vehicle controller, vehicle control method, and vehicle control program

Abstract

To diminish the frequency of performing a steering manipulation, which a driver feels tiresome, in driving assist control (lane sustaining control).SOLUTION: A vehicle controller (DS) implements driving assist control (lane sustaining control) of automatically steering an own vehicle so that a lateral position of the own vehicle relative to a travel lane will coincide with a predetermined target lateral position. When a scene where the own vehicle is traveling is a predefined learning scene, if a driver performs a steering manipulation, the vehicle controller stores a value, which depends on the steering manipulation, in association with the learning scene. Thereafter, when a scene specified with ambient information shifts to the learning scene during implementation of driving assist control, and / or, when a scene specified with ambient information is the learning scene at a time when implementation of driving assist control is initiated, the vehicle controller varies the target lateral position (including the deviation initiation timing) on the basis of the stored learning value.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a vehicle control device, a vehicle control method, and a vehicle control program that automatically steers a vehicle so that the lateral position of the vehicle coincides with a target lateral position. [Background technology]

[0002] One conventional vehicle control device (hereinafter referred to as the "conventional device") sets the center line of the vehicle's lane (driving lane) as a target driving line, and executes lane keeping control to automatically steer the vehicle so that the reference point of the vehicle moves along the target driving line. Furthermore, when the driver's steering force becomes equal to or exceeds a threshold value while executing lane keeping control, the conventional device changes the position of the target driving line so that the target driving line becomes the line between the center line and the reference point of the vehicle (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-62261 A Summary of the Invention

[0004] For example, in a particular scene where the vehicle is traveling in the vehicle's own lane in a temporary shared section, the driver may wish to always drive the vehicle along "not the center line of the vehicle's own lane, but a line slightly shifted from the center line in the direction approaching the road shoulder." However, with the conventional device, such a driver must apply a large steering force to the steering wheel every time a particular scene starts, and finds such steering operation troublesome. The present invention has been made to solve such a problem.

[0005] One aspect of the present invention is An information acquisition device (10, 20, 30) that acquires surrounding information that indicates a surrounding situation of the host vehicle, including a lane marking of a lane in which the host vehicle is traveling; a controller (10) that executes driving assistance control (e.g., lane keeping control) that automatically steers the host vehicle based on the peripheral information so that the lateral position of the host vehicle with respect to the driving lane coincides with a target lateral position set within the driving lane; Equipped with The controller: When a steering deviation state occurs in which the lateral position of the host vehicle deviates from the target lateral position due to a steering operation by the driver of the host vehicle during execution of the driving assist control (S240, S525), if the scene identified by the peripheral information is a predetermined learning scene (first scene, second scene), a value corresponding to the steering operation by the driver is associated with the learning scene and stored as a learning value (S245, S555), Furthermore, after storing the learning value (S225, S450), the controller changes the target lateral position based on the stored learning value (S255, S480) in at least one of the following cases: when the scene specified by the peripheral information changes from a scene other than the learning scene to the learning scene during execution of the driving assistance control (S220, S440), and when the scene specified by the peripheral information at the time when execution of the driving assistance control is started is the learning scene (S220); It is structured as follows.

[0006] According to the above aspect, when the scene specified by the peripheral information is a predetermined learning scene and a steering deviation state occurs in which the lateral position of the vehicle deviates from the target lateral position due to the driver's steering operation, a value corresponding to the steering operation is associated with the learning scene and stored. Then, when the scene specified by the peripheral information changes from a scene other than the learning scene to the learning scene during the execution of the driving assistance control, and / or when the scene specified by the peripheral information at the time when the execution of the driving assistance control is started is the learning scene, the target lateral position is changed based on the stored learning value. As a result, the target lateral position becomes a value according to the driver's preference, so that the possibility that the driver has to perform a steering operation every time a similar scene occurs is reduced. Therefore, the frequency of steering operations that the driver finds annoying can be reduced.

[0007] In addition, in order to accurately determine whether a scene specified by the peripheral information is a predetermined learning scene using image data, the vehicle control device may perform supervised training. Furthermore, the present invention also covers a vehicle control method and a program thereof. Each of the constituent elements of the present invention is not limited to the embodiments defined by the symbols in the above summary of the invention. [Brief description of the drawings]

[0008] [Figure 1] 1 is a schematic configuration diagram of a vehicle control device according to a first embodiment of the present invention. [Diagram 2] 2 is a flowchart showing a routine executed by the vehicle control ECU of FIG. 1. [Diagram 3] FIG. 2 is a diagram for explaining the operation of the vehicle control device according to the first embodiment of the present invention. [Figure 4] 10 is a routine (flowchart) executed by a vehicle control ECU (second ECU) of a vehicle control device according to a second embodiment of the present invention. [Diagram 5] 4 is a flowchart showing a routine executed by a second ECU. [Figure 6]FIG. 6 is a diagram for explaining the operation of a vehicle control device according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] A vehicle control device DS (hereinafter referred to as "device DS") according to a first embodiment of the present invention has components shown in Fig. 1 and described below, and is applied to (mounted on) a vehicle. A vehicle control device according to a second embodiment of the present invention, described below, also has a similar configuration to the device DS.

[0010] In this specification, an "ECU" is an electronic control unit having a microcomputer including a CPU (processor) and memory as its main part, and is also called a controller. The components shown in Fig. 1, including multiple ECUs, are connected to each other via a CAN (Controller Area Network) so that they can exchange information with each other.

[0011] The vehicle control ECU 10 executes lane keeping control, which is one type of driving assistance control to be described later.

[0012] The front camera device 20 includes a front camera 21 and an image ECU 22. The front camera 21 captures an image of a scene in front of the vehicle (including white and yellow markings, and targets such as poles and other vehicles) at every predetermined time to obtain image data. The image ECU 22 generates camera target information including the position, relative longitudinal speed, relative lateral speed, and type of target in the image data by analyzing the image data from the front camera 21. The target types include poles, roadside structures, other vehicles, and the like. Furthermore, the image ECU 22 obtains lateral position information including the lateral position of the vehicle in the lane in which the vehicle is traveling (hereinafter also referred to as the "travel lane") (the distance between the center line of the left and right markings in the lane width direction and the reference point of the vehicle) and the angle (yaw angle) between the traveling direction of the vehicle and the center line, based on the image data. Furthermore, the image ECU 22 identifies the type of the marking line (white line, yellow line, etc.) based on the image data and obtains the marking line identification result. The image ECU 22 transmits the obtained information to the vehicle control ECU 10.

[0013] The radar device 30 is a well-known device that acquires information about a target present in front of the vehicle using millimeter-wave band radio waves, and includes a radar 31 and a radar ECU 32. The radar 31 transmits information about the transmitted radio waves and the received reflected waves to the radar ECU 32. The radar ECU 32 acquires radar target information based on the information from the radar 31, and transmits it to the vehicle control ECU 10. The radar target information includes the distance to the target, the direction of the target, the relative speed of the target, etc.

[0014] The powertrain ECU 35 is connected to a powertrain actuator 36 so as to be capable of driving the powertrain actuator 36. The actuator 36 can change the torque generated by a drive device (drive power source) of the host vehicle. The torque generated by the drive device is transmitted to the drive wheels via a gear mechanism. Therefore, the powertrain ECU 35 can control the drive power of the host vehicle.

[0015] If the drive device of the host vehicle is a gasoline-fueled internal combustion engine, the actuator 36 is, for example, a throttle valve actuator that changes the opening of a throttle valve. The host vehicle may be an electric vehicle, in which case the actuator 36 is an inverter that can change the torque of an electric motor. The host vehicle may be a hybrid vehicle, in which case the actuator 36 includes an inverter for the electric motor and a throttle valve actuator for the internal combustion engine.

[0016] The brake ECU 40 drives the brake actuator 41 to control the friction braking devices provided on each wheel of the host vehicle, thereby changing the friction braking force applied to the host vehicle. That is, the brake ECU 40 can control the braking force applied to the host vehicle.

[0017] The steering ECU 50 drives a steering motor 51 to control the steering angle of the host vehicle.

[0018] The navigation ECU 60 is connected to a GPS (or GNSS) receiver 61, a map database 62, and a display touch panel 63 that displays touch buttons. The navigation ECU 60 acquires the current position of the vehicle based on the GPS signal received by the GPS receiver 61. The navigation ECU 60 acquires information about the road on which the vehicle is currently traveling from the map database 62. The road information includes, for example, information about whether the lane on which the vehicle is currently traveling is a vehicle-only road and whether it is within a provisional shared section. Note that a provisional shared section refers to a section of a road that is originally two lanes on one side (four lanes in total) but has been provisionally changed to one lane on one side (two lanes in total).

[0019] The vehicle control ECU 10 is further connected to "sensors that detect parameters indicating the driving state of the vehicle" and switches, which will be described below, and is adapted to input values ​​(signals) detected or output by these sensors. A vehicle speed sensor 71 for detecting the speed of the host vehicle (i.e., vehicle speed SPD). An accelerator pedal operation amount sensor 72 that detects the operation amount AP of the accelerator pedal of the host vehicle. A brake pedal operation amount sensor 73 detects an operation amount BP of the brake pedal of the host vehicle. A steering angle sensor 74 that detects a steering angle θ of the host vehicle. A steering torque sensor 75 that detects the steering torque Tq input to the steering wheel of the host vehicle. A lane keeping control operation switch 76 that is operated by the driver of the vehicle and outputs either an ON signal (lane keeping control request signal) or an OFF signal (lane keeping control end request signal).

[0020] (Operation of the first embodiment) The CPU of the vehicle control ECU 10 (hereinafter simply referred to as the "CPU") executes a routine shown in the flowchart of FIG. 2 every time a predetermined time elapses.

[0021] When an appropriate time arrives, the CPU starts processing from step 200 (hereinafter, "step" will be abbreviated as "S") in FIG. 2 and proceeds to S205, where it analyzes the image data to identify what type of scene is in front of the vehicle.

[0022] Next, the CPU proceeds to S210 and determines whether the CPU is currently executing lane keeping control. The CPU executes a routine (not shown) to determine whether the lane keeping control execution condition is satisfied, and during the period when the lane keeping control execution condition is satisfied, executes a well-known lane keeping control that controls the steering angle based on lateral position information so that a reference point of the host vehicle (e.g., a center position between the left front wheel and the right front wheel) coincides with a predetermined target lateral position.

[0023] The lane keeping control execution condition is met, for example, when both of the following conditions 1 and 2 are met, and is not met when at least one of the conditions 1 and 2 is not met. (Condition 1) The road on which the vehicle is traveling is a road reserved for vehicles. Whether or not Condition 1 is established is determined based on the analysis result of the image data in S205 and / or the road information obtained from the navigation ECU 60. (Condition 2) The signal from the lane keeping control operation switch 76 operated by the driver is an ON signal. That is, the driver is requesting the execution of lane keeping control.

[0024] If the CPU is currently executing lane keeping control, the CPU proceeds from S210 to S215, where it determines whether the value of the temporary shared section flag Xz is "0".

[0025] If the value of the provisional shared section flag Xz is "0" (i.e., the driving lane has been recognized as being outside the provisional shared section up to the present time), the CPU proceeds from S215 to S220, and determines whether the present time is immediately after the start of the provisional shared section based on the analysis result of the image data in S205. In other words, the CPU determines whether the driving lane has entered the provisional shared section from outside the provisional shared section in S220. That is, the CPU determines whether the scene specified by the surrounding information including the image data has changed to "a first scene indicating that the driving lane is in the provisional shared section". Note that, when the CPU recognizes that the center line of the driving lane is "yellow line, white line, and pole" in the order from the driving lane side to the oncoming lane, it determines that the driving lane is in the provisional shared section. The CPU may obtain information on whether the driving lane is in the provisional shared section from the map database 62 and perform the determination in S220.

[0026] If the current time is just after the start of the temporary shared section, the CPU proceeds from S220 to S225, and determines whether the value of the lateral position learned flag Xg (described later) is "1" to determine whether the host vehicle has traveled through the temporary shared section before and has stored (learned) the lateral position of the host vehicle at that time as a learned value (first learned value).

[0027] If the value of the lateral position learned flag Xg is not "1", the CPU proceeds directly from S225 to S230, and sets the value of the provisional shared section flag Xz to "1". After that, the CPU proceeds to S295, and ends this routine. In this case, the target lateral position is set to the reference lateral position (a default value, for example, the center position between the left and right lane lines). As a result, the next time the CPU proceeds to S215, the CPU proceeds from S215 to S235.

[0028] In S235, the CPU determines whether the current time is immediately after the temporary shared section ends. In other words, in S235, the CPU determines whether the driving lane has moved from inside the temporary shared section to outside the temporary shared section. That is, the CPU determines whether the scene specified by the peripheral information including the image data has changed from the first scene to a scene other than the first scene. If the current time is immediately after the temporary shared section ends, the CPU proceeds from S235 to S240.

[0029] In S240, the CPU determines whether or not a steering deviation state has occurred during travel in the provisional shared section, in which the lateral position of the vehicle deviates by a predetermined distance or more from the "target lateral position set at the start of the provisional shared section" due to the driver's steering operation. If a steering deviation state has occurred, the CPU proceeds to S245, where it stores the average value of the lateral position of the vehicle in the most recent predetermined time in a non-volatile memory (not shown) as the target lateral position learning value (first learning value) of the provisional shared section, and sets the value of the lateral position learned flag Xg to "1." Next, the CPU proceeds to S250, where it sets the value of the provisional shared section flag Xz to "0," and then proceeds to S295. On the other hand, if a steering deviation state has not occurred, the CPU proceeds directly from S240 to S250.

[0030] Furthermore, if the current time is not immediately after the temporary shared section ends, the CPU proceeds from S235 to S255 to determine whether the current time is "immediately before the lane keeping control is terminated due to the lane keeping control execution condition not being satisfied." If the current time is immediately before the lane keeping control is terminated, the CPU proceeds from S255 to S240. On the other hand, if the lane keeping control is currently continuing, the CPU proceeds from S255 to S295.

[0031] When the processing of S245 and S250 is executed, the value of the lateral position learned flag Xg is set to "1". If the CPU then proceeds to S225, the process proceeds from S225 to S255, and the target lateral position is set to the target lateral position learned value (first learned value) of the provisional shared section.

[0032] If lane keeping control is not being performed when the CPU proceeds to S210, the CPU proceeds directly from S210 to S295 and ends this routine for the time being. If the driving lane is in a temporary shared section when lane keeping control is started, the CPU judges "Yes" in each step from S210 to S220 and proceeds to S225.

[0033] Since the device DS operates as described above, for example, when the host vehicle enters the provisional shared section during lane keeping control at time t1 in FIG. 3A and learning of the first learned value has not been completed, the target lateral position is set to the reference lateral position (the position of the center line between the left and right lane lines). The reference lateral position is a position where the distance DL from the left lane line is D1 and the distance DR from the right lane line is D1. After that, the driver steers the vehicle during the period from time t2 to time t3, and the position where the host vehicle is traveling after time t3 shifts to the left of the center line. Then, when the lane keeping control is completed or the provisional shared section ends at time t4, the device DS stores "the lateral position where the distance DL is D3 and the distance DR is D2" as the first learned value.

[0034] After that, for example, when lane keeping control is started at time t5 in FIG. 3B and the device DS recognizes that the driving lane at that time is entering the temporary shared section, the target lateral position is set to the first learned value (the lateral position where the distance DL is D3 and the distance DR is D2). After that, the driver steers the vehicle during the period from time t6 to time t7, and the position where the host vehicle is traveling shifts to the right after time t7. Then, when the lane keeping control ends or the temporary shared section ends at time t8, the device DS stores "the lateral position where the distance DL is D5 and the distance DR is D4" as the first learned value.

[0035] (Operation of the second embodiment) The CPU of the vehicle control ECU 10 according to the second embodiment executes the routine shown in the flowcharts of Figures 4 and 5 every time a predetermined time elapses. Thus, when an appropriate time arrives, the CPU starts the process from S400 in Figure 4 and proceeds to S410, where it is determined whether or not the CPU is currently executing lane keeping control.

[0036] If the CPU is currently executing lane keeping control, the CPU proceeds from S410 to S420, where it analyzes the image data to identify the type of scene in front of the vehicle. Next, in S430, the CPU determines whether the value of the variable lateral position control flag XL is "0." If the variable lateral position control (VLO: Variable Lateral Offset), which will be described later, is being executed, the value of the variable lateral position control flag XL is set to "1," and if the variable lateral position control is not being executed, the value of the variable lateral position control flag XL is set to "0."

[0037] When the value of the variable lateral position control flag XL is "0", the CPU proceeds from S430 to S440, and determines whether or not the current time has just been recognized as "the second scene in which the host vehicle HV overtakes the adjacent vehicle AV" based on the analysis result in S420, as shown in Fig. 6(A). The second scene starts when "an adjacent vehicle AV, which is another vehicle traveling at a lower vehicle speed than the host vehicle HV" is traveling in the adjacent lane AL adjacent to the driving lane HL, and the margin time TTC until the front end of the host vehicle HV comes closest to the rear end of the adjacent vehicle AV (the margin time TTC = (the distance between the adjacent vehicle AV and the host vehicle HV) / the relative speed of the adjacent vehicle AV)) becomes equal to or less than the second scene setting threshold.

[0038] If the current time is not immediately after it has been recognized as the second scene, the CPU proceeds directly from S440 to S495 and ends this routine for the time being. On the other hand, if the current time is immediately after it has been recognized as the second scene, the CPU proceeds from S440 to S450 and determines whether the value of the change timing learned flag Xt is "1" or not, thereby determining whether the host vehicle has stored (learned) the lateral position change timing as a learned value (second learned value) up to the current time.

[0039] If the value of the learned flag Xt is not "1", the CPU proceeds from S450 to S460 and sets the lateral position change timing, which indicates the threshold margin time, to the reference timing (a default value whose unit is time, for example, 2 seconds). Next, the CPU sets the value of the variable lateral position control flag XL to "1" in S470 and proceeds to S495.

[0040] On the other hand, if the value of the learned flag Xt is "1", the CPU proceeds from S450 to S480, and sets the lateral position change timing, which indicates the threshold margin time, to a timing learned value (second learned value) described later. Next, the CPU proceeds to S470 and S495.

[0041] If the CPU determines "No" in S410 (i.e., lane keeping control is not being executed), the CPU proceeds directly to S495. If the CPU determines "No" in S430 (i.e., variable lateral position control is being executed), the CPU proceeds directly to S495.

[0042] When an appropriate time arrives, the CPU starts the process from S500 in FIG. 5 and proceeds to S505, where it determines whether or not the CPU is currently executing lane keeping control.

[0043] If the CPU is currently executing lane keeping control, the CPU proceeds from S505 to S510, where it analyzes the image data to identify the type of scene ahead of the vehicle. Next, in S515, the CPU determines whether the value of the variable lateral position control flag XL is "1".

[0044] If the value of the variable lateral position control flag XL is "1", the CPU proceeds from S515 to S520 and determines whether the above-mentioned second scene is being continuously recognized based on the analysis result in S510. If the second scene is being continuously recognized, the CPU proceeds from S520 to S525 and determines whether the steering torque Tq is equal to or greater than a threshold torque, thereby determining whether the driver has performed a steering operation (i.e., whether a steering deviation state has occurred).

[0045] If the driver does not perform a steering operation, the CPU proceeds from S525 to S530 to determine whether the actual approach margin time TTC is equal to or less than the threshold margin time set as the lateral position change timing. If the approach margin time TTC is not equal to or less than the lateral position change timing, the CPU proceeds directly from S530 to S540. In contrast, if the approach margin time TTC is equal to or less than the lateral position change timing, the CPU proceeds from S530 to S535 to start shifting the target lateral position by the variable lateral position change control. As a result, during the period from when the approach margin time TTC is equal to or less than the lateral position change timing to when the variable lateral position control end condition described later is satisfied, the target lateral position is shifted by a predetermined distance D in a direction away from the adjacent lane in which the adjacent vehicle is traveling (see time t11 to time t13 in FIG. 6A).

[0046] In S540, the CPU determines whether the value of the variable lateral position control flag XL is "1." If the value of the variable lateral position control flag XL is "0," the CPU proceeds directly from S540 to S595 and ends this routine for the time being. On the other hand, if the value of the variable lateral position control flag XL is "1," the CPU proceeds from S540 to S545 and determines whether an end condition for the variable lateral position control is satisfied. This end condition is satisfied when it is determined that the second scene has disappeared or that the host vehicle has overtaken an adjacent vehicle based on "a rear camera image acquired by a rear camera (not shown) or target information acquired by a rear side radar (not shown)."

[0047] If the variable lateral position control end condition is not satisfied, the CPU proceeds directly from S545 to S595. On the other hand, if the variable lateral position control end condition is satisfied, the CPU proceeds from S545 to S550, sets the value of the variable lateral position control flag XL to "0", and then proceeds to S595.

[0048] On the other hand, when the CPU proceeds to S525, if the driver performs a steering operation in a direction to move the vehicle away from the adjacent lane before the approach margin time TTC becomes equal to or less than the lateral position change timing, the CPU proceeds from S525 to S555, where it learns (stores) the approach margin time at that time as a second learning value (timing learning value). Next, the CPU proceeds to S560, where it sets the value of the change timing learned flag Xt to "1". After that, the CPU proceeds to S535. In this case, the driving line of the vehicle changes in response to the driver's steering operation as shown by the dashed line from time t10 in FIG. 6(A) when the driver performed the steering operation.

[0049] As a result, when the CPU next recognizes a scene in which the vehicle overtakes an adjacent vehicle (second scene), the process proceeds from S450 to S480 in Fig. 4, and the lateral position change timing is set to the timing learning value (second learning value). Therefore, the timing at which the determination is "Yes" in S530 in Fig. 5 changes to a timing that satisfies the driver's preference (see time t20 and thereafter in Fig. 6(B)).

[0050] If the CPU determines "No" in each of steps S505, S515, and S520, it proceeds directly to S540 from each step.

[0051] As described above, according to the device DS, when the driver performs a steering operation when the scene in which the vehicle is traveling is a predetermined learning scene, a value corresponding to the steering operation is stored in association with the learning scene. Then, when the scene identified by the peripheral information changes to the learning scene during the execution of the driving assistance control thereafter and / or when the scene identified by the peripheral information at the time when the execution of the driving assistance control is started is the learning scene, the target lateral position (including the shift start timing) is changed based on the stored learning value. As a result, the target lateral position becomes a value according to the driver's preference, so that the possibility that the driver has to perform a steering operation every time a similar scene occurs is reduced.

[0052] The present invention is not limited to the above-mentioned embodiment and modified examples, and various modified examples can be adopted within the scope of the present invention. For example, the device DS may execute lane keeping control using the above-mentioned learning value as lane keeping control during automatic driving. The device DS may execute variable lateral position change control in a scene where an oncoming vehicle passes by another vehicle, and in that case, the timing at which the target lateral position starts to shift due to the variable lateral position change control (the relative relationship between the oncoming vehicle and the vehicle when a steering operation is performed) is learned. Furthermore, for example, as in the device disclosed in Patent Document 1, control may be performed to change the target lateral position when a steering deviation state occurs due to a steering operation while the lane keeping control is being performed. In that case, for example, the target lateral position at the time when the driving lane leaves the temporary shared section may be learned as the first learning value. In addition, the learning scene is not limited to the above-mentioned example, and may be, for example, a scene of a one-lane road with a roadside structure (for example, a high soundproof wall) installed on the center median or road shoulder, and a scene where a cone indicating that construction is underway is installed on the road shoulder side of the driving lane. [Explanation of symbols]

[0053] 10...vehicle control ECU, 20...front camera device, 30...radar device.

Claims

1. an information acquisition device that acquires surrounding information representing a surrounding situation of the host vehicle, the surrounding information including a lane marking of a lane in which the host vehicle is traveling; a controller that executes driving assistance control to automatically steer the host vehicle based on the peripheral information so that a lateral position of the host vehicle with respect to the driving lane coincides with a target lateral position set within the driving lane; Equipped with The controller: when a steering deviation state occurs in which a lateral position of the host vehicle deviates from the target lateral position due to a steering operation by a driver of the host vehicle during execution of the driving assist control, if a scene identified by the peripheral information is a predetermined learning scene, a value corresponding to the steering operation of the driver is associated with the learning scene and stored as a learning value, Furthermore, after storing the learned value, the controller When a scene identified by the peripheral information changes from a scene other than the learning scene to the learning scene during execution of the driving assistance control, And, When the scene identified by the surrounding information at the time when the execution of the driving assistance control is started is the learning scene, In at least one of the above, the target lateral position is changed based on the stored learning value. It was configured as follows: Vehicle control device.

2. The vehicle control device according to claim 1, The controller: When the steering deviation state occurs in a case where the scene identified by the peripheral information during execution of the driving assistance control is a first scene as the learning scene indicating that the driving lane is in a temporary shared section, a position corresponding to an actual lateral position of the vehicle as a value corresponding to a steering operation of the driver is stored as a first learned value that is one of the learned values, Furthermore, after storing the first learning value, the controller When a scene specified by the peripheral information changes from a scene other than the first scene to the first scene during execution of the driving assistance control, And, When the scene identified by the surrounding information at the time when the execution of the driving assistance control is started is the first scene, In at least one of the cases, and executing the driving assistance control by setting the target lateral position to a value based on the stored first learned value. It was configured as follows: Vehicle control device.

3. The vehicle control device according to claim 1, The controller: When a scene specified by the surrounding information during execution of the driving assistance control is a second scene as the learning scene in which the host vehicle travels within the driving lane and overtakes an adjacent vehicle traveling in a lane adjacent to the driving lane, when a relative relationship between the host vehicle and the adjacent vehicle reaches a predetermined relationship, a variable lateral position control is executed as the driving assistance control to shift the target lateral position by a predetermined distance in a direction away from the adjacent vehicle, Further, the controller when the steering deviation state occurs before the target lateral position starts to be shifted by the variable lateral position control, a relative relationship between the host vehicle and the adjacent vehicle is stored in association with the second scene as a second learned value, which is one of the learned values; and when a scene specified by the peripheral information changes from a scene other than the second scene to the second scene during execution of the driving assistance control after the second learning value is stored, when a relative relationship between the host vehicle and the adjacent vehicle becomes a relationship determined based on the second learning value, a shift of the target lateral position in the variable lateral position control is started. Vehicle control device.

4. acquiring surrounding information representing a surrounding situation of the host vehicle including a lane marking of a lane in which the host vehicle is traveling; executing a driving assistance control for automatically steering the host vehicle based on the peripheral information so that a lateral position of the host vehicle with respect to the driving lane coincides with a target lateral position set within the driving lane; A vehicle control method comprising: when a steering deviation state occurs in which a lateral position of the host vehicle deviates from the target lateral position due to a steering operation by a driver of the host vehicle during execution of the driving assist control, if a scene identified by the peripheral information is a predetermined learning scene, storing a value corresponding to the steering operation of the driver as a learning value in association with the learning scene; After storing the learned value, When a scene identified by the peripheral information changes from a scene other than the learning scene to the learning scene during execution of the driving assistance control, And, When the scene identified by the surrounding information at the time when the execution of the driving assistance control is started is the learning scene, In at least one of the above, changing the target lateral position based on the stored learned value; A vehicle control method comprising:

5. A program to be executed by a computer mounted on a vehicle, The program includes: acquiring surrounding information representing a surrounding situation of the host vehicle including a lane marking of a lane in which the host vehicle is traveling; executing a driving assistance control for automatically steering the host vehicle based on the peripheral information so that a lateral position of the host vehicle with respect to the driving lane coincides with a target lateral position set within the driving lane; Run the command, The program further causes the computer to when a steering deviation state occurs in which a lateral position of the host vehicle deviates from the target lateral position due to a steering operation by a driver of the host vehicle during execution of the driving assist control, if a scene identified by the peripheral information is a predetermined learning scene, storing a value corresponding to the steering operation of the driver as a learning value in association with the learning scene; After storing the learned value, When a scene identified by the peripheral information changes from a scene other than the learning scene to the learning scene during execution of the driving assistance control, And, When the scene identified by the surrounding information at the time when the execution of the driving assistance control is started is the learning scene, In at least one of the above, changing the target lateral position based on the stored learned value; A program that executes the following.

Images