Vehicle control device

Abstract

To provide a vehicle control device that can properly learn liking of a driver for an inter-vehicle distance of an own vehicle with respect to a preceding vehicle when the driver stops the own vehicle during manual operation.SOLUTION: The vehicle control device controls an own vehicle that enables a driver to switch between manual operation and automatic operation. The vehicle control device comprises a processor and a storing device. The processor executes learning control by which an inter-vehicle distance of the own vehicle with respect to a preceding vehicle is learnt when the driver stops the own vehicle during the manual operation and reflects a learnt result of the inter-vehicle distance by the leaning control on control of an inter-vehicle distance during the automatic operation. The storing device stores the inter-vehicle distance as learning data during the manual operation. In the learning control, the processor determines whether the storing device is made to store the inter-vehicle distance as the learning data or not, on the basis of information on vehicle operation by the driver in a series of periods of time from a period of time before the own vehicle is stopped to a period of time after the own vehicle is stopped.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present disclosure relates to a vehicle control device capable of switching between manual driving and automatic driving. [Background technology]

[0002] Patent Document 1 discloses a driving characteristic learning method. In the driving characteristic learning method, in a vehicle that can switch between manual driving by a driver and automatic driving, the inter-vehicle distance during deceleration operation by the driver when manually driving is prioritized for learning. In addition, in the driving characteristic learning method, a learning condition for determining whether the current driving state is suitable for acquiring data to be used for learning the driving characteristics includes that the inter-vehicle distance from the preceding vehicle when the vehicle is stopped is within a predetermined value. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2018 / 138767 Summary of the Invention [Problem to be solved by the invention]

[0004] It is conceivable to execute a learning control that learns the driver's preference regarding the distance between the host vehicle and a preceding vehicle when the driver stops the host vehicle during manual driving, and to reflect the learning results in the control of the distance between the host vehicle and a preceding vehicle during automatic driving. Regarding the distance between the host vehicle and a preceding vehicle when the vehicle is stopped, Patent Document 1 discloses only that the distance between the host vehicle and a preceding vehicle is within a predetermined value as a learning condition for the driving characteristics. However, the learning condition disclosed in this way alone is not sufficient to appropriately learn the driver's preference regarding the distance between the host vehicle and a preceding vehicle when the vehicle is stopped, which is the target of the learning control.

[0005] The present disclosure has been made in consideration of the above-mentioned problems, and aims to provide a vehicle control device that can appropriately learn the driver's preferences regarding the distance between the vehicle and a preceding vehicle when the driver stops the vehicle during manual driving. [Means for solving the problem]

[0006] A vehicle control device according to the present disclosure controls a host vehicle capable of switching between manual driving by a driver and autonomous driving. The control device includes a processor and a storage device. The processor executes learning control to learn the following distance of the host vehicle relative to a preceding vehicle when the driver stops the host vehicle during manual driving, and reflects the learning result of the learning control on the following distance in the control of the following distance during autonomous driving. The storage device stores the following distance as learning data during manual driving. In the learning control, the processor determines whether to store the following distance as learning data in the storage device based on vehicle operation information by the driver during a series of periods from a period before stopping the host vehicle to a period after stopping the host vehicle. Additionally, machine learning may be used to acquire the learning result of the following distance through the learning control. [Effects of the Invention]

[0007] According to the present disclosure, it is possible to appropriately learn the driver's preferences regarding the distance between the host vehicle and a preceding vehicle when the driver stops the host vehicle during manual driving. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a diagram schematically illustrating an example of the configuration of a vehicle (host vehicle) according to an embodiment. [Figure 2] 10 is a table showing an example of a list of learning conditions C according to the embodiment. [Figure 3] FIG. 4 is a diagram showing a specific example of the type of vehicle of a preceding vehicle when the host vehicle is stopped. [Figure 4] 4 is a flowchart illustrating an example of a process related to vehicle driving control according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0010] 1. Vehicle configuration 1 is a diagram schematically illustrating an example of the configuration of a vehicle 1 according to an embodiment. The vehicle 1 is equipped with a vehicle control system 10. The vehicle control system 10 is mounted on the vehicle 1 and controls the driving of the vehicle 1. The vehicle control system 10 includes a vehicle state sensor 12, a recognition sensor 14, a position sensor 16, a communication device 18, a driving device 20, an electronic control unit (ECU) 22, and a driving selector switch 24.

[0011] The vehicle state sensor 12 detects the state of the vehicle 1. The vehicle state sensor 12 includes, for example, a vehicle speed sensor, an acceleration sensor, an accelerator pedal sensor, a brake pedal sensor, a steering angle sensor, and a turn signal sensor. The turn signal sensor detects the operation state of the turn signal. The recognition sensor 14 recognizes (detects) the situation around the vehicle 1. The recognition sensor 14 includes, for example, a camera. The position sensor 16 detects the position and orientation of the vehicle 1. The position sensor 16 includes, for example, a GNSS (Global Navigation Satellite System) receiver.

[0012] The communication device 18 communicates with the outside of the vehicle 1. The communication device 18 communicates with, for example, an external system to acquire various information. The information includes, for example, map information and traffic information. The map information includes road information such as road gradient. The traffic information includes, for example, congestion information, information on road surface conditions, and information on visibility such as fog or heavy rain. The communication device 18 may also include, for example, a vehicle-to-vehicle communication device that enables communication between the vehicle 1 and surrounding vehicles (i.e., vehicle-to-vehicle communication (V2V)).

[0013] The traveling device 20 is a device that operates the vehicle 1. For example, the traveling device 20 includes a drive device, a braking device, and a steering device. The drive device includes, for example, at least one of an electric motor and an internal combustion engine for driving (accelerating) the vehicle 1. The braking device includes a brake actuator for braking (deceleration) the vehicle 1. The steering device includes, for example, a steering motor for steering the vehicle 1.

[0014] The ECU 22 is a computer that controls the vehicle 1, and corresponds to an example of a "vehicle control device" according to the present disclosure. The ECU 22 includes a processor 26 and a storage device 28. The processor 26 executes various processes, including processes related to vehicle driving control, which will be described later. The storage device 28 stores various information necessary for the processes performed by the processor 26. The various processes performed by the ECU 22 are realized by the processor 26 executing a computer program. The computer program is stored in the storage device 28. Alternatively, the computer program may be recorded on a computer-readable recording medium. The ECU 22 may be configured by combining multiple ECUs.

[0015] The vehicle control system 10 is configured to be able to execute automatic driving control that controls "automatic driving" of the vehicle 1. This automatic driving control includes vehicle stop control that controls the inter-vehicle distance D so that the vehicle (host vehicle) 1 stops behind a preceding vehicle at a target inter-vehicle distance Dt. More specifically, the automatic driving referred to here corresponds to, for example, level 3 or higher automatic driving as defined by the Society of Automotive Engineers (SAE) in the United States, but is not necessarily limited to level 3 or higher automatic driving. In other words, the automatic driving control may be any control that includes the vehicle stop control described above, and may be, for example, adaptive cruise control (ACC) that can perform braking control down to 0 km / h. Known technologies are applied to such automatic driving control. Therefore, a detailed description of the automatic driving control will be omitted.

[0016] The driving mode changeover switch 24 is operated by the driver and is used to switch the driving mode of the vehicle 1 between manual driving by the driver and automatic driving. In other words, the vehicle 1 is configured to be switchable between manual driving and automatic driving. Manual driving is performed by the driver operating the accelerator pedal, brake pedal, and steering wheel at his / her own will.

[0017] 2. Vehicle driving control In this embodiment, when the driver operates the driving mode selector switch 24 to select automatic driving, the ECU 22 executes automatic driving control including the vehicle stop control described above. On the other hand, when the driver operates the manual driving mode, the ECU 22 executes the following "vehicle distance learning control."

[0018] 2-1. Distance learning control when stopped The inter-vehicle distance learning control (or simply learning control) according to this embodiment learns the inter-vehicle distance D (see, for example, FIGS. 3(A) to 3(C)) of the host vehicle 1 relative to the preceding vehicle when the driver stops the host vehicle 1 during manual driving. The learning result of the inter-vehicle distance D by this learning control (i.e., the learning value DL described below) is reflected in the control of the inter-vehicle distance D during automatic driving.

[0019] The learning condition C is a condition for executing learning by the learning control. In other words, the learning condition C is a condition for executing storage of the inter-vehicle distance D, which is learning data in the learning control. In order to accurately capture the driver's preference regarding the inter-vehicle distance D when stopped, the learning condition C is set as follows:

[0020] FIG. 2 is a table showing an example of a list of learning conditions C according to an embodiment. In the example shown in FIG. 2, the learning condition C includes a prerequisite C0 and three types of learning conditions C1 to C3. The learning condition C1 is a learning condition related to the pre-stop period P1 (pre-stop learning condition). The learning condition C2 is a learning condition related to the stop process P2 (stop process condition). The learning condition C3 is a learning condition related to the post-stop period P3 (post-stop learning condition). The learning condition C is established when all of these conditions C0 to C3 are satisfied.

[0021] The pre-stop period P1 corresponds to the period before deceleration based on the driver's operation to stop the host vehicle 1 begins. The pre-stop period P1 may be, for example, the period from the end of the post-stop period P3 related to the previous stop of the host vehicle 1 to the start time t0 of deceleration (including braking) for the current stop. Alternatively, if the start time t0 of the deceleration can be predicted, the pre-stop period P1 may be a predetermined period ending at the start time t0 of the deceleration. The stopping process P2 is the period from the start time t0 of deceleration to the completion time t1 of stopping the host vehicle 1. The post-stop period P3 is a predetermined period from the completion time t1 of stopping.

[0022] The learning condition C includes learning conditions C1 to C3, and thus covers a series of periods P, including a pre-stop period P1, a stopping process P2, and a post-stop period P3. By using these learning conditions C1 to C3, in the learning control according to this embodiment, the ECU 22 (processor 26) determines whether or not to store the inter-vehicle distance D as learning data in the storage device 28, based on vehicle operation information by the driver during the series of periods P.

[0023] First, the pre-stop learning condition C1 includes, for example, that the vehicle speed V of the host vehicle 1 reaches a threshold value or more after the previous stop (condition C1-1). In other words, condition C1-1 requires that the host vehicle 1 is not traveling slowly during the pre-stop period P1. When the host vehicle 1 stops from a state where it is traveling slowly due to factors such as the location of a traffic light ahead of the host vehicle 1 or traffic congestion, the inter-vehicle distance D tends to depend on the surrounding conditions. As a result, the acquired inter-vehicle distance D does not accurately capture the driver's preferences. Therefore, according to condition C1-1, scenes in which the host vehicle 1 stops after traveling slowly, i.e., scenes in which the driver's preferences are difficult to accurately capture, can be excluded from the learning target. Note that there may be multiple pre-stop learning conditions C1.

[0024] Next, the stopping process learning condition C2 includes, for example, conditions C2-1 to C2-5. Condition C2-1 is that the host vehicle 1 stops after the preceding vehicle stops. Condition C2-2 is that both the preceding vehicle and the host vehicle 1 stop while traveling straight ahead. Condition C2-3 is that the host vehicle 1 does not change lanes during the stopping process P2. Condition C2-4 is that the host vehicle 1 stops facing the same preceding vehicle (in other words, that no other vehicle has cut in or crossed between the preceding vehicle and the host vehicle 1). Condition C2-5 is that the driver of the host vehicle 1 has no intention of turning right or left or overtaking the preceding vehicle. By using these conditions C2-1 to C2-5, an inter-vehicle distance D that is easy for the driver to control can be captured as learning data.

[0025] Furthermore, the learning condition C2 includes, for example, the following: the inter-vehicle distance D obtained at the time of the current stop is equal to or less than a threshold value (condition C2-6). This is because if the inter-vehicle distance D is long, there is a possibility that a large inter-vehicle distance D was maintained in order to yield to another vehicle. The learning condition C2 also includes, for example, the stopping of the host vehicle 1 not being accompanied by sudden braking by the driver (condition C2-7). This is because it is unlikely that the host vehicle 1 has stopped at an inter-vehicle distance D that meets the driver's preferences. The learning condition C2 also includes, for example, the braking being performed by the driver's own operation (condition C2-8). This is because the driver's preferences can be captured by excluding braking due to an emergency brake that is automatically applied under control by the vehicle control system 10 during manual driving.

[0026] Next, the post-stop learning condition C3 includes, for example, that both the preceding vehicle and the host vehicle 1 have been stopped for a predetermined period of time (condition C3-1). The reason for this is to capture a stable stopping scene. This predetermined period may be the same as or shorter than the predetermined period that specifies the post-stop period P3. Furthermore, the learning condition C3 includes, for example, that the driver of the host vehicle 1 has no intention of turning right or left or overtaking the preceding vehicle (condition C3-2). The reason for this is that there is a possibility that the driver may be maintaining a different inter-vehicle distance D than usual in order to turn right or left or overtake.

[0027] Furthermore, the learning condition C3 includes, for example, that the stop of the host vehicle 1 is not for parking (condition C3-3). This is because the following distance D may differ from the driver's preference in order to fit the host vehicle 1 into a parking space. The learning condition C3 also includes, for example, that no other vehicle cuts in or crosses between the preceding vehicle and the host vehicle 1 (condition C3-4). This is because the driver may have secured a large following distance D in order to yield to the other vehicle. The learning condition C3 also includes, for example, that the host vehicle 1 has not changed the following distance D after stopping (condition C3-5). This is because it is difficult to determine whether the following distance D before or after the change reflects the driver's preference. Note that the post-stop learning condition C3 may include, for example, any one or more of conditions C3-1 to C3-5.

[0028] Next, the prerequisite C0 includes conditions C0-1 and C0-2, which are conditions based on the vehicle type of the preceding vehicle. Here, the inter-vehicle distance D adjusted by the driver of the host vehicle 1 during manual driving is affected by the height H of the preceding vehicle. That is, if the height H is large, the driver's forward visibility is likely to be obstructed while the host vehicle 1 is stopped. For example, it becomes difficult to see traffic lights or signs ahead. For this reason, when stopping behind a preceding vehicle with a large height H, the driver is likely to maintain a longer inter-vehicle distance D than when stopping behind a preceding vehicle with a similar or smaller height H. In this way, the inter-vehicle distance D may vary depending on the height H of the preceding vehicle.

[0029] In view of the above, condition C0-1 is that the height H of the preceding vehicle is less than a threshold value TH1. That is, in learning control, if the height H of the preceding vehicle is less than a predetermined threshold value TH1, ECU 22 stores the inter-vehicle distance D for the preceding vehicle as learning data in storage device 28. On the other hand, if the height H is equal to or greater than threshold value TH1, ECU 22 does not store the inter-vehicle distance D as learning data in storage device 28. In this way, by limiting the learning targets to vehicle types that are easy for the driver of host vehicle 1 to maintain visibility, it is possible to avoid variations in the learning data for the inter-vehicle distance D due to differences in vehicle types.

[0030] 3(A) to 3(C) are diagrams showing specific examples of the type of preceding vehicle when the host vehicle 1 is stopped. In the example shown in FIG. 3(A), the preceding vehicle 2 has a height H1 equivalent to that of the host vehicle 1. On the other hand, in the example shown in FIG. 3(B), the preceding vehicle 3 has a height H2 greater than that of the host vehicle 1.

[0031] Specifically, the preceding vehicle 2 is an automobile, and more specifically, for example, a four-wheeled automobile. The preceding vehicle 2 corresponds to, for example, a standard automobile (a standard passenger car, a compact passenger car, or a light automobile) under the Road Traffic Act of Japan. In addition, the preceding vehicle 2 may be a private vehicle or a commercial vehicle. On the other hand, the preceding vehicle 3 is a larger automobile (such as a truck or a bus) than the preceding vehicle 2. More specifically, the preceding vehicle 3 corresponds to, for example, a medium-sized automobile, a medium-sized automobile, or a large automobile under the Road Traffic Act of Japan.

[0032] The threshold value TH1 is determined in advance as a value that can distinguish between the preceding vehicle 2 and the preceding vehicle 3. That is, the threshold value TH1 is located between the height H1 of the preceding vehicle 2 and the height H2 of the preceding vehicle 3. In other words, the threshold value TH1 corresponds to a value that can distinguish a standard-sized vehicle from a vehicle with a height H greater than that of a standard-sized vehicle.

[0033] In the example described with reference to preceding vehicles 2 and 3, it is assumed that host vehicle 1 is a standard automobile, like preceding vehicle 2. However, the vehicle type of host vehicle 1 is not necessarily limited to a standard automobile. In other words, the condition for excluding a preceding vehicle from the learning target in consideration of its vehicle type may be, for example, that the height H of the preceding vehicle is higher than the height of host vehicle 1, and that the relative height ΔH of the preceding vehicle with respect to host vehicle 1 is equal to or greater than a predetermined threshold.

[0034] 3(C), the preceding vehicle 4 is a two-wheeled vehicle (motorcycle). Here, when the host vehicle 1 is a four-wheeled vehicle, if the preceding vehicle is a two-wheeled vehicle, it is considered that the driver tends to have difficulty in grasping the appropriate inter-vehicle distance D due to the size, behavior, etc. of the preceding vehicle. For this reason, when following a two-wheeled vehicle, it is considered that the driver tends to try to maintain a longer inter-vehicle distance D, or the inter-vehicle distance D becomes unstable, compared to when following a four-wheeled vehicle such as a standard car.

[0035] In view of the above, condition C0-2 is that the preceding vehicle is not a two-wheeled vehicle. That is, the conditions for excluding the preceding vehicle from the learning target in consideration of the vehicle type include the preceding vehicle being a two-wheeled vehicle. Therefore, in the learning control, if the preceding vehicle is a two-wheeled vehicle, ECU 22 does not store the inter-vehicle distance D as learning data in storage device 28. This makes it possible to avoid variations in the learning data for inter-vehicle distance D due to differences in the sense of proximity to the preceding vehicle caused by differences in vehicle type.

[0036] As described above, according to conditions C0-1 and C0-2, whether or not to store learning data (inter-vehicle distance D) for a certain preceding vehicle in storage device 28 is determined (changed) depending on the vehicle type of the preceding vehicle. In addition, instead of conditions C0-1 and C0-2, for example, one of the prerequisites C0 may include that the vehicle type of host vehicle 1 and the preceding vehicle are the same.

[0037] Furthermore, the prerequisite C0 includes, for example, conditions C0-3 to C0-8. Condition C0-3 is that the road on which the host vehicle 1 is traveling is not congested. Condition C0-4 is that the road surface and visibility conditions are good (condition C0-4). Condition C0-5 is that the host vehicle 1 is not off the road (for example, in a parking lot or on a premises). Condition C0-6 is that the road on which the host vehicle 1 is traveling is neither a residential road nor a narrow street. Condition C0-7 includes that the traveling point of the host vehicle 1 is not a junction of multiple roads. Condition C0-8 is that the traveling point of the host vehicle 1 is not a toll booth. According to each of these conditions C0-3 to C0-8, it is possible to avoid situations in which the driver's following distance D is likely to be based on the surrounding conditions, and to accurately capture the driver's preferences.

[0038] Furthermore, the precondition C0 includes, for example, a condition that the absolute value of the road gradient is small (condition C0-9). The reason for this is that if the absolute value of the road gradient is large, such as on a steep downhill slope, the driver will be careful of the host vehicle 1 sliding downhill when stopped, and the driver's preferences may not be reflected in the inter-vehicle distance D. The precondition C0 also includes, for example, a condition that the host vehicle 1 is not traveling on a motorway (expressway) (condition C0-10). The reason for this is that the host vehicle 1 stops on a motorway due to factors such as traffic congestion, and the inter-vehicle distance D is likely to depend on the surrounding conditions.

[0039] 2-2. Processing flow 4 is a flowchart showing an example of processing related to vehicle driving control according to the embodiment. The processing of this flowchart is repeatedly executed while the vehicle control system 10 is running.

[0040] In step S100, the ECU 22 (processor 26) determines whether the vehicle 1 is being manually driven or automatically driven. This determination can be made based on the operation state of the driving mode changeover switch 24, for example.

[0041] If the vehicle 1 is being manually driven (step S100; Yes), the process proceeds to step S102. In step S102, the ECU 22 acquires various pieces of information necessary for executing learning control. The various pieces of information include the inter-vehicle distance D when stopped. More specifically, if the host vehicle 1 is stopped behind the preceding vehicle (i.e., the vehicle speed V is 0 km / h) when the process proceeds to step S102, the various pieces of information are acquired, for example, using the recognition sensor 14. The various pieces of information also include information necessary for determining whether the learning condition C (see, for example, FIG. 2) is satisfied. The information includes the height H of the preceding vehicle. The height H is acquired, for example, using the recognition sensor 14 or vehicle-to-vehicle communication. Other information for determining the learning condition C is also acquired, for example, using information from the vehicle state sensor 12 and the recognition sensor 14, road information (map information), and traffic information.

[0042] In step S104 following step S102, the ECU 22 determines whether the learning condition C is satisfied. More specifically, it determines whether all of the above-described conditions C0 to C3 are satisfied. As a result, if the learning condition C is satisfied, that is, if it is determined that the stopping scene is one in which the driver's preference is reflected in the inter-vehicle distance D, the process proceeds to step S106. In addition, the learning condition C is satisfied when all of the conditions C0 to C3 are satisfied at the timing when the above-described post-stop period P3 has elapsed.

[0043] On the other hand, if at least one of the above conditions C0 to C3 is not satisfied, the learning condition C is not established. If the learning condition C is not established, the process proceeds to RETURN.

[0044] In step S106, the ECU 22 stores the inter-vehicle distance D acquired in step S102. That is, if the learning condition C is met, the inter-vehicle distance D is stored in the storage device 28 as learned data.

[0045] In step S108 following step S106, the ECU 22 determines whether a predetermined number of data items for the inter-vehicle distance D have been accumulated. This predetermined number is, for example, 30. If the predetermined number of data items for the inter-vehicle distance D have not yet been accumulated (step S108; No), the process proceeds to Return. On the other hand, if the predetermined number of data items have been accumulated (step S108; Yes), the process proceeds to step S110.

[0046] In step S110, the ECU 22 calculates a learned value DL of the inter-vehicle distance D. For example, the ECU 22 calculates the average value of the accumulated data of the inter-vehicle distance D as the learned value DL. Thereafter, the process proceeds to step S112.

[0047] In step S112, the ECU 22 applies the learned value DL calculated in step S110 to the target inter-vehicle distance Dt. That is, the target inter-vehicle distance Dt is updated with the latest learned value DL. The target inter-vehicle distance Dt is used in vehicle stop control performed during autonomous driving.

[0048] On the other hand, if the vehicle 1 is in autonomous driving (step S100; No), the process proceeds to step S114. In step S114, the ECU 22 determines whether a predetermined vehicle stop condition involving a preceding vehicle is satisfied, for example, using information obtained from the recognition sensor 14. As a result, if the vehicle stop condition is satisfied, the process proceeds to step S116. On the other hand, if the vehicle stop condition is not satisfied, the process proceeds to return.

[0049] In step S116, the ECU 22 controls the (actual) inter-vehicle distance D so as to realize the target inter-vehicle distance Dt by controlling the traveling device 20. In this way, by the processing of steps S112 to S116, the learning result (learned value DL) of the inter-vehicle distance D by the learning control is reflected in the control of the inter-vehicle distance D during automatic driving.

[0050] In addition, acquisition of the learning value DL may be performed using, for example, a machine learning model instead of the process shown in FIG. 4 . That is, this machine learning model is constructed, for example, with various predetermined parameters as input and the learning value DL as output. The various parameters are, for example, multiple parameters related to the above-mentioned precondition C0. More specifically, a parameter indicating that each precondition C0 is satisfied (for example, that the road on which the host vehicle 1 is traveling is not congested) and a parameter indicating that each precondition C0 is not satisfied (for example, that the road is congested) can be used. The machine learning model is trained, for example, using training data acquired while the vehicle 1 is traveling (i.e., the above-mentioned various parameters as explanatory variables (input) and the inter-vehicle distance D as the objective variable).

[0051] 3.Effects The driver's operation when stopping the host vehicle 1 during manual driving is affected not only by the stopping process P2 but also by various environments and situations around the host vehicle 1 during a series of periods P, including a pre-stop period P1 and a post-stop period P3 (see, for example, the description of FIG. 2 ). These influences are reflected in the inter-vehicle distance D at the time of stopping based on the driver's operation. According to the present embodiment described above, whether or not to store the inter-vehicle distance D as learning data in the storage device 28 is determined based on vehicle operation information by the driver during the series of periods P. As described above, according to the present embodiment, the vehicle operation by the driver during the series of periods P, including before and after stopping, and the vehicle behavior associated with the operation, are taken into account as the learning condition C. This reduces the variation in the learning results due to the influence of various environments and situations around the host vehicle 1 during the series of periods P. Therefore, the driver's preferences regarding the inter-vehicle distance D can be learned at an appropriate time. Then, during autonomous driving, the target inter-vehicle distance Dt appropriately learned based on the driver's preferences can be used. This makes it possible to realize autonomous driving with reduced driver stress.

[0052] 4. Other examples of learning control Each of the above-described preconditions C0-1 to C0-6, C0-9, and C0-10 may be used as follows, instead of limiting the driving scene to be learned as in the process of step S104. That is, for example, with respect to precondition C0-3, the learning control of this embodiment may be executed separately for when the road on which the host vehicle 1 is traveling is not congested and when the road is congested. The same applies to the other preconditions C0-1, C0-2, C0-4 to C0-6, C0-9, and C0-10. [Explanation of symbols]

[0053] 1 vehicle, 2, 3, 4 preceding vehicle, 10 vehicle control system, 12 vehicle state sensor, 14 recognition sensor, 16 position sensor, 18 communication device, 20 running device, 22 electronic control unit (ECU), 24 driving changeover switch, 26 processor, 28 storage device

Claims

1. A control device for controlling a vehicle that can switch between manual driving by a driver and automatic driving, a processor that executes learning control to learn the inter-vehicle distance of the host vehicle relative to a preceding vehicle when the driver stops the host vehicle during the manual driving, and reflects the learning result of the inter-vehicle distance by the learning control in the control of the inter-vehicle distance during the automatic driving; a storage device that stores the inter-vehicle distance as learned data during the manual driving; Equipped with In the learning control, the processor determines whether to store the inter-vehicle distance as the learning data in the storage device based on vehicle operation information by the driver during a series of periods from a pre-stop period to a post-stop period of the host vehicle; a learning condition for storing the learning data in the learning control includes a pre-stop learning condition related to the pre-stop period, The pre-stop learning condition includes that the vehicle speed of the host vehicle has reached a threshold value greater than 0 after the previous stop. Vehicle control device.

2. A control device for controlling a vehicle that can switch between manual driving by a driver and automatic driving, a processor that executes learning control to learn the inter-vehicle distance of the host vehicle relative to a preceding vehicle when the driver stops the host vehicle during the manual driving, and reflects the learning result of the inter-vehicle distance by the learning control in the control of the inter-vehicle distance during the automatic driving; a storage device that stores the inter-vehicle distance as learned data during the manual driving; Equipped with In the learning control, the processor determining whether to store the inter-vehicle distance as the learned data in the storage device based on vehicle operation information by the driver during a series of periods from a pre-stop period to a post-stop period of the host vehicle; If the height of the preceding vehicle is less than a threshold value, the inter-vehicle distance is stored in the storage device as the learned data; If the height is equal to or greater than the threshold value, the inter-vehicle distance is not stored in the storage device as the learned data. Vehicle control device.

3. A control device for controlling a vehicle that can switch between manual driving by a driver and automatic driving, a processor that executes learning control to learn the inter-vehicle distance of the host vehicle relative to a preceding vehicle when the driver stops the host vehicle during the manual driving, and reflects the learning result of the inter-vehicle distance by the learning control in the control of the inter-vehicle distance during the automatic driving; a storage device that stores the inter-vehicle distance as learned data during the manual driving; Equipped with In the learning control, the processor determines whether to store the inter-vehicle distance as the learning data in the storage device based on vehicle operation information by the driver during a series of periods from a pre-stop period to a post-stop period of the host vehicle; The host vehicle is a four-wheeled automobile, In the learning control, when the preceding vehicle is a two-wheeled vehicle, the processor does not store the inter-vehicle distance as the learning data in the storage device. Vehicle control device.

4. a learning condition for storing the learning data in the learning control includes a post-stop learning condition related to the post-stop period, The post-stop learning condition is: The stopped state of both the preceding vehicle and the host vehicle continues for a predetermined period of time. The driver has no intention to turn right or left or to overtake the preceding vehicle; The stop of the vehicle is not for parking; No other vehicle has cut in or crossed between the preceding vehicle and the host vehicle, and The vehicle-to-vehicle distance has not been changed after the vehicle has stopped. Contains at least one of The vehicle control device according to any one of claims 1 to 3.

Images