WO2023002899A1

WO2023002899A1 - Driving assistance device and driving assistance method

Info

Publication number: WO2023002899A1
Application number: PCT/JP2022/027609
Authority: WO
Inventors: 真司北浦
Original assignee: 株式会社デンソー
Priority date: 2021-07-23
Filing date: 2022-07-13
Publication date: 2023-01-26
Also published as: JPWO2023002899A1

Abstract

A driving assistance device (5) comprises a path storage unit (54) that stores, in a storage unit (50), as stored information, path information including a driving route of a vehicle and the surrounding conditions of the vehicle on the driving route, which are successively obtained by a recognition processing unit (51) when an exiting operation or entering operation for the vehicle has been performed. The driving assistance device comprises: a path generation unit (55) that generates a target route that the vehicle should follow when the vehicle exits or enters, such generation being on the basis of the stored information stored by the storage unit. If route information, newly acquired by the recognition processing unit after storing the stored information in the storage unit, contains information different from the stored information, the path storage unit updates the stored information with the different information.

Description

Driving support device, driving support method

Cross-references to related applications

This application is based on Japanese Patent Application No. 2021-121134 filed on July 23, 2021, the contents of which are incorporated herein by reference.

The present disclosure relates to a driving support device and a driving support method.

　Conventionally, as a parking assistance device, a device that automatically parks a vehicle in a predetermined parking space is known (see Patent Document 1, for example). The parking assistance device described in Patent Document 1 stores information including a travel route from a reference start position to a parking position and the surroundings of the travel route when the vehicle is parked by a driver's operation, and stores the information in the information. Based on this, a parking route that is a target route during parking assistance is generated.

Japanese Patent Publication No. 2013-530867

However, the parking assistance device described in Patent Document 1 stores the information such as the driving route only at the timing instructed by the driver. , it becomes difficult to generate an appropriate target path.
An object of the present disclosure is to provide a driving assistance device and a driving assistance method capable of improving robustness against environmental changes.

According to one aspect of the present disclosure,
The driving support device is
Stored information includes the driving route of the vehicle and the surrounding conditions of the vehicle on the driving route, which are successively acquired by the recognition processing unit when the driver leaves the vehicle from the parking space or enters the parking space. A route storage unit that stores in the storage unit as
a route generation unit that generates a target route that the vehicle should follow when leaving or entering the vehicle based on the stored information stored in the storage unit;
The route storage unit updates the stored information with different information when the route information newly acquired by the recognition processing unit after the storage information is stored in the storage unit includes information different from the stored information.

According to another aspect of the disclosure,
The driving support method is
Stored information includes the driving route of the vehicle and the surrounding conditions of the vehicle on the driving route, which are successively acquired by the recognition processing unit when the driver leaves the vehicle from the parking space or enters the parking space. and storing in the storage unit as
generating a target route that the vehicle should follow when leaving or entering the vehicle based on the stored information stored in the storage unit;
The route storage unit updates the stored information with different information when the route information newly acquired by the recognition processing unit after the storage information is stored in the storage unit includes information different from the stored information.

According to these, even if the environment of the parking space changes after the stored information is stored in the storage unit, the stored information is updated by the route information newly acquired by the recognition processing unit. It is possible to improve robustness against

It should be noted that the reference numerals in parentheses attached to each component etc. indicate an example of the correspondence relationship between the component etc. and the specific component etc. described in the embodiment described later.

1 is a schematic configuration diagram of an automatic parking system according to an embodiment; FIG. FIG. 10 is an explanatory diagram for explaining a situation when the vehicle is parked facing backward; FIG. 10 is an explanatory diagram for explaining an example of a travel route when leaving the garage facing forward; It is a flow chart which shows a flow of learning processing which a vehicle control part of an embodiment performs. FIG. 10 is an explanatory diagram for explaining an example of a travel route of a vehicle when obstacles increase in the vicinity of a parking space; 4 is a flow chart showing a flow of route update executed by a vehicle control unit; 4 is a flow chart showing the flow of target update performed by a vehicle control unit; 4 is a flowchart showing the flow of follow-up control processing executed by a vehicle control unit; It is an explanatory view for explaining the situation when leaving the parking lot facing backwards. FIG. 10 is an explanatory diagram for explaining an example of a travel route when the vehicle is parked in a forward direction;

An embodiment of the present disclosure will be described below with reference to FIGS. 1 to 8. FIG. In this embodiment, an example in which the driving assistance device 5 and the driving assistance method of the present disclosure are applied to the automatic driving system 1 will be described. As shown in FIG. 1 , the automatic driving system 1 has surrounding monitoring sensors 3 , various ECUs 4 , and a driving support device 5 . The driving support device 5 is communicably connected to the perimeter monitoring sensor 3 and various ECUs 4 directly or via an in-vehicle LAN (Local Area Network).

The surroundings monitoring sensor 3 is an autonomous sensor that monitors the surroundings of the vehicle V itself. For example, the surroundings monitoring sensor 3 detects an obstacle OB composed of three-dimensional objects around the own vehicle, such as moving dynamic targets such as pedestrians and other vehicles, and stationary static targets such as structures on the road. Also, a parking assist mark indicating parking information, which is information about the parking lot PL, is detected as a detection target. Here, as the surroundings monitoring sensor 3, a surroundings monitoring camera 31 that captures a predetermined range around the vehicle, a sonar 32 that transmits search waves to a predetermined range around the vehicle, a millimeter wave radar 33, and a LiDAR (Light Detection and Ranging) sensor. A probe wave sensor such as 34 is provided.

The surroundings monitoring camera 31 corresponds to an image capturing device, captures an image of the surroundings of the own vehicle, and outputs the imaged data to the driving support device 5 as sensing information. Here, the front camera 31a, the rear camera 31b, the left side camera 31c, and the right side camera 31d, which capture images in front of, behind, and on the left and right sides of the vehicle, are exemplified as the peripheral monitoring camera 31, but the camera is limited to this. not a thing

The search wave sensor outputs the search wave and acquires the reflected wave, and the measurement results such as the relative speed and relative distance to the target and the azimuth angle at which the target exists are used as sensing information for the driving support device. 5 sequentially. The sonar 32 performs measurement using ultrasonic waves as survey waves, and is provided at a plurality of locations on the vehicle V. For example, a plurality of sonars 32 are arranged side by side in the left and right direction of the vehicle on the front and rear bumpers, and are arranged around the vehicle. Measurement is performed by outputting a probe wave. The millimeter wave radar 33 performs measurement using millimeter waves as search waves. The LiDAR 34 performs measurements using laser light as the probe wave. Both the millimeter wave radar 33 and the LiDAR 34 output search waves, for example, within a predetermined range in front of the vehicle V, and perform measurement within that output range.

In the present embodiment, the peripheral monitoring sensor 3 includes a peripheral monitoring camera 31, a sonar 32, a millimeter-wave radar 33, and a LiDAR 34. One or more of these sensors can be used for peripheral monitoring. It's fine if you can do it, but you don't have to have everything.

In addition, the driving support device 5 is connected to the map database 35 and the GPS 36, and is capable of measuring the current position of its own vehicle V. Note that the map database 35 may be built in the vehicle-mounted device, or may be built in the device outside the vehicle (for example, an external server).

The driving support device 5 constitutes an ECU (that is, an electronic control device) for performing various controls for realizing the driving support method in the automatic driving system 1, and includes a CPU, a storage unit 50, I / O, etc. It is composed of a microcomputer with

The storage unit 50 includes ROM, RAM, EEPROM, and the like. That is, the storage unit 50 has a volatile memory 50a such as RAM and a nonvolatile memory 50b such as EEPROM. The storage unit 50 is composed of a non-transition tangible recording medium.

The driving support device 5 supports the driving operation of the vehicle V by the driver between the public road OL and the parking space SP provided on the land along the public road OL. As shown in FIG. 2, the driving assistance device 5 of the present embodiment assists the operation of entering the vehicle V into the parking space SP. Although FIG. 2 shows the situation when the vehicle is entering the parking lot facing backward, the driving support by the driving support device 5 is not limited to the case of entering the parking space facing backward, but can also be performed when entering the parking space facing forward. can be done.

The driving support device 5 controls the vehicle V when the vehicle V is parked based on the route information including the travel route of the vehicle V when the driver enters or leaves the vehicle V and the surrounding conditions of the vehicle V on the travel route. Generate a target path that V should follow. During driving support, the driving support device 5 automatically moves the vehicle V along the target route to the parking space SP, which is the planned parking position, or automatically moves the vehicle V from the parking space SP to a predetermined position. let

For example, as shown in FIG. 3 , the driving support device 5 detects the travel route of the vehicle V sequentially acquired by the recognition processing unit 51 and the vehicle V on the travel route when the driver leaves the vehicle V. route information including the surrounding situation is stored in the storage unit 50 as storage information. For example, as shown in FIG. 2, the driving support device 5 stores the route information sequentially acquired by the recognition processing unit 51 when the driver enters the vehicle V into the storage unit 50 as a storage state. It may be configured to memorize. In this way, the driving support device 5 performs learning processing for storing route information during manual driving by the driver. This learning process is executed, for example, when a learning switch (not shown) is operated by the driver. The "surrounding situation of the vehicle V" includes, for example, dynamic objects such as people and other vehicles around the vehicle V, curbs around the vehicle V, static objects such as buildings, various signs, and guide lines. This is the situation of road markings, etc.

While moving from the learning start position STP, which is the start position of the learning process, to the learning stop position SPP, the driving assistance device 5 detects targets, parkable objects, etc. Recognize free space, parking position, etc. These recognition results are stored in the storage unit 50 as route information and used for driving assistance.

The driving support device 5 generates a target route and performs various controls for driving support based on the information stored in the storage unit 50 and the route information sequentially acquired by the recognition processing unit 51 during driving support. That is, when an instruction for driving assistance is issued, the driving assistance device 5 generates a target route based on the sensing information stored in the storage unit 50 and the sensing information of the surrounding monitoring sensor 3 during driving assistance, and Perform route tracking control according to. Specifically, the driving support device 5 includes a recognition processing unit 51, a vehicle information acquisition unit 52, and a vehicle control unit 53 as functional units that execute various controls.

The recognition processing unit 51 receives sensing information from the surroundings monitoring sensor 3, and based on the sensing information, recognizes the surrounding environment of the vehicle that is about to be parked, recognizes the scene in which parking is to be performed, Furthermore, it recognizes three-dimensional objects existing around the vehicle. Here, the recognition processing section 51 is composed of an image recognition section 51a, a space recognition section 51b, a free space recognition section 51c, and a map information collation section 51d.

The image recognition unit 51a has a scene recognition unit 51aa, a three-dimensional object recognition unit 51ab, and a landmark/marker recognition unit 51ac. They input imaging data from the surrounding monitoring camera 31 as sensing information and perform image analysis of the imaging data to perform scene recognition, three-dimensional object recognition, and landmark/marker recognition.

In scene recognition, what kind of scene the parking scene is, for example, whether it is a normal parking scene with sufficient width of the path to the parking space SP, or whether the path to the parking space SP is obstructed by an obstacle OB Recognize whether it is a special parking scene with no room in the width.

The imaging data input from the surroundings monitoring camera 31 shows the surroundings, so if the image is analyzed, it can be determined whether it is a normal parking scene or a special parking scene. For example, the frontage of the parking space SP is calculated from the imaging data, and if the frontage is within a predetermined range preset based on the width of the own vehicle, it can be determined as a special parking scene. Note that the scene recognition may be performed based on not only the sensing information of the perimeter monitoring camera 31 but also the sensing information of the survey wave sensor.

In three-dimensional object recognition, obstacles OB composed of three-dimensional objects that exist around the vehicle, such as dynamic targets and static targets, are recognized as objects to be detected. Based on the detection object recognized by this three-dimensional object recognition, preferably the shape of a static target among them, the above-described scene recognition and the generation of the driving support map including the obstacle OB are performed.

In landmark/marker recognition, landmarks indicating parking spaces SP, graphic marks indicating parking lots PL, graphic marks indicating parking directions, two-dimensional barcodes indicating parking scenes, parking directions, etc. are recognized. The scene recognition described above is performed based on the marks recognized by the landmark marker recognition. Also, for example, a parking assist mark indicating parking information such as a parking scene is attached in or in front of the parking space SP. can recognize the type of parking scene indicated by .

The space recognition unit 51b also has a three-dimensional object recognition unit 51ba. The three-dimensional object recognition unit 51ba recognizes three-dimensional objects in the space around the vehicle based on sensing information from at least one of the sonar 32, millimeter wave radar 33, and LiDAR34. The three-dimensional object recognition here is the same as the three-dimensional object recognition performed by the image recognition section 51a. Therefore, if either one of the image recognition section 51a and the space recognition section 51b is provided, three-dimensional object recognition can be performed. In the case of the present embodiment, the space recognition unit 51b does not perform scene recognition, but the space recognition unit 51b performs scene recognition based on sensing information from at least one of the sonar 32, the millimeter wave radar 33, and the LiDAR 34. can also be done.

Although three-dimensional object recognition and scene recognition can be performed by either the image recognition unit 51a or the space recognition unit 51b, it is possible to perform three-dimensional object recognition and scene recognition with higher accuracy by using both. . For example, by complementing the three-dimensional object recognition and scene recognition by the image recognition unit 51a with the three-dimensional object recognition and scene recognition by the space recognition unit 51b, it is possible to perform three-dimensional object recognition and scene recognition with higher accuracy.

The free space recognition unit 51c performs free space recognition to recognize a free space in the parking lot PL. The free space means, for example, a space of a size and shape that allows the vehicle V to run in the parking lot PL.

The free space recognition unit 51c recognizes free spaces in the parking lot PL based on the recognition results of scene recognition and three-dimensional object recognition by the image recognition unit 51a and the space recognition unit 51b. For example, from the results of scene recognition and three-dimensional object recognition, the shape of the parking lot PL and the presence or absence of parking of other vehicles can be grasped, so based on this, the free space in the parking lot PL is recognized.

The map information collating unit 51d acquires map information from the map database 35 and collates the acquired map information with the position information acquired by the GPS 36 to identify the position of the vehicle V on the map as the current position. The map information collating unit 51d may specify the current position of the vehicle V using information other than the map information of the map database 35 and the position information acquired by the GPS 36 .

The vehicle information acquisition unit 52 acquires information on the amount of operation of the vehicle V from other ECUs 4 and the like. Specifically, the vehicle information acquisition unit 52 acquires detection signals output from sensors such as an accelerator position sensor, a brake depression force sensor, a steering angle sensor, a wheel speed sensor, and a shift position sensor mounted on the vehicle V. .

The vehicle control unit 53 executes various controls required for driving assistance. Specifically, the vehicle control unit 53 includes a route storage unit 54, a route generation unit 55, a position estimation unit 56, and a follow-up control unit 57 as functional units that execute various controls.

The route storage unit 54 stores a travel route through which the vehicle V passes when the driver performs one of the operation of exiting the parking space SP and the operation of entering the parking space SP, and the surrounding conditions of the vehicle V on the travel route. is stored in the storage unit 50 . The route storage unit 54 of the present embodiment stores route information in the storage unit 50 when the driver performs an operation to leave the parking space SP. Specifically, the route storage unit 54 stores, as storage information, route information such as targets on the travel route of the vehicle V, free parking spaces, and parking positions, which are sequentially acquired by the recognition processing unit 51, in the storage unit 50. do.

Here, when the route information newly acquired by the recognition processing unit 51 after the storage information is stored in the storage unit 50 includes information different from the stored information, the route storage unit 54 updates the stored information with the different information. . More specifically, the route storage unit 54 stores the route information acquired by the recognition processing unit 51 when the driver performs the leaving operation or the entering operation after storing the stored information in the storage unit 50 . If the information is different from that stored in the property memory 50a, the stored information is updated. Further, the route storage unit 54 includes information different from the stored information in the route information newly acquired by the recognition processing unit 51 when the vehicle V automatically leaves or enters the garage by the follow-up control unit 57, which will be described later. If so, update the stored information with different information. Here, updating is a concept that includes addition of new route information to the storage unit 50, deletion of part of the route information stored in the storage unit 50, and the like, in addition to changing the route information. For example, updating information includes not only overwriting previously acquired information with newly acquired information, but also adding newly acquired information without deleting previously acquired information. is included.

The route generation unit 55 generates a route based on the results of scene recognition, three-dimensional object recognition, and free space recognition. The route generation unit 55 generates a target route that the vehicle V should follow when entering or leaving the vehicle V based on the route information stored during the learning process. For example, the route generation unit 55 sets the travel route of the vehicle V as a reference route, and if there is a section in the reference route in which the distance between the vehicle V and the obstacle OB is equal to or less than a predetermined value, the section is defined as the vehicle V and the obstacle OB. A target route is generated by replacing it with a route whose distance from the OB exceeds a predetermined value. The obstacle OB is composed of a three-dimensional object recognized by three-dimensional object recognition.

Here, the route generation unit 55 avoids the obstacles OB when generating the target route, but may generate the target route by avoiding only static targets among them. Since the dynamic target moves, the vehicle V can be moved only after the risk of collision with the dynamic target has disappeared. It is enough to generate

The position estimation unit 56 estimates the current position of the vehicle V based on the stored information stored in the storage unit 50 and the route information sequentially acquired by the recognition processing unit 51 during driving assistance. The position estimation unit 56 compares, for example, the sensing information stored in the storage unit 50 and the sensing information acquired during driving assistance, and estimates the current position based on the difference between them.

The follow-up control unit 57 automatically moves the vehicle V to the parking space SP along the target route by performing vehicle motion control such as acceleration/deceleration control and steering control of the vehicle V. Specifically, the follow-up control unit 57 outputs control signals to various ECUs 4 so that the vehicle V reaches the parking space SP along the target route.

The various ECUs 4 include a steering ECU 41 that controls steering, a brake ECU 42 that controls acceleration and deceleration, a power management ECU 43, and a body ECU 44 that controls various electrical components such as lights and door mirrors.

Specifically, the follow-up control unit 57 outputs from each sensor such as an accelerator position sensor, a brake depression force sensor, a steering angle sensor, a wheel speed sensor, and a shift position sensor mounted on the vehicle V via the vehicle information acquisition unit 52. A detected signal is obtained. The follow-up control unit 57 detects the state of each unit from the acquired detection signal, and outputs control signals to various ECUs 4 in order to move the vehicle V following the target route.

Here, there may be static targets that were not recognized when the target route was first calculated. Therefore, three-dimensional object recognition by the three-dimensional object recognition units 51ab and 51ba is continued even while the vehicle V is moving following the target route. Further, if a static target exists in a place where the vehicle V may collide with the vehicle V following the target route, the target route is regenerated. For example, when moving the vehicle V from the current position to the parking space SP, a person or another vehicle may approach. For this reason, the follow-up control unit 57 temporarily stops the vehicle V until the dynamic target comes out of the range of the planned movement trajectory of the vehicle V estimated from the target route and the vehicle width. to avoid colliding with dynamic targets.

The follow-up control unit 57 includes a proposal unit 58 that proposes a target route to the driver. The proposal unit 58 makes various proposals including driving assistance to the driver through the HMI (abbreviation of Human Machine Interface) 45 . The HMI 45 is a device for providing various types of support to the driver. HMI 45 functions as a notification device having an ECU, a display, a speaker, and the like. Note that the proposal unit 58 may not be a function specialized for the follow-up control unit 57, and may be used together with other function units.

The automatic driving system 1 according to this embodiment is configured as described above. Next, the operation of the automatic driving system 1 configured in this manner will be described. In this embodiment, as shown in FIG. 2, the case where the vehicle V is parked in the parking space SP provided in the parking lot PL along the public road OL will be described as an example. In the parking lot PL shown in FIG. 2, there is a portion having a passage width through which one vehicle V can pass in the passage from the public road OL to the parking space SP. In addition, FIG. 2 illustrates backward parking in which the vehicle V is moved backward and parked in the parking space SP.

First, the learning process for storing route information during manual driving by the driver will be described with reference to the flowchart shown in FIG. The learning process shown in FIG. 4 is executed by the vehicle control unit 53 at predetermined control cycles when an instruction to start the vehicle from a parked state is issued, such as when the ignition switch IG is turned on. Each processing shown in this flowchart is implemented by each functional unit of the driving support device 5 . In addition, each step for realizing this processing can also be grasped as each step for realizing the driving support method.

As shown in FIG. 4, the vehicle control unit 53 starts recognition processing in step S100. In this recognition process, sensing information from the periphery monitoring sensor 3 is input, and the above-described scene recognition, three-dimensional object recognition, and free space recognition are started based on the input sensing information. Also, in the recognition process, detection of the current position of the vehicle V using the map database 35 and GPS 36 is started.

Subsequently, in step S110, the vehicle control unit 53 determines whether or not the current position of the vehicle V is the learning start position STP specified in advance by the driver. For example, the parking space SP in front of the home is set as the learning start position STP. Note that the learning start position STP is not limited to the position designated by the driver, and may be, for example, the start position or stop position of the previous driving assistance. Also, the learning start position STP may be determined from the current recognition result based on the previous learning results. The current recognition result includes, for example, information output from at least one of the recognition processing unit 51, the map database 35, the GPS 36, other map information, and the like.

If the current position of the vehicle V is the learning start position STP, the vehicle control unit 53 proceeds to step S120, and if the current position of the vehicle V is not the learning start position STP, the vehicle control unit 53 skips the subsequent processes and exits the learning process. .

At step S120, the vehicle control unit 53 starts storing various information necessary for driving assistance. The vehicle control unit 53 stores the route information including the traveling route of the vehicle V and the surrounding conditions of the traveling route successively acquired by the recognition processing unit 51 as storage information in the storage unit 50 .

Subsequently, in step S130, the vehicle control unit 53 determines whether or not the current position of the vehicle V is the learning stop position SPP designated in advance by the driver. The learned stop position SPP is set, for example, at the vehicle entrance/exit B that is the boundary between the public road OL and the parking lot PL. Note that the learning stop position SPP is not limited to the position designated by the driver, and may be, for example, the start position or stop position of the previous driving assistance. Also, the learning stop position SPP may be determined from the current recognition result based on the previous learning results. The current recognition result includes, for example, information obtained from the output of the recognition processing unit 51, the map database 35, the GPS 36, other map information, and the like.

The vehicle control unit 53 waits when the current position of the vehicle V is not the learned stop position SPP, and proceeds to step S140 when the current position of the vehicle V is the learned stop position SPP. In step S140, the vehicle control unit 53 stops storing various information necessary for driving assistance.

Subsequently, in step S150, the vehicle control unit 53 determines whether or not the environment around the parking space SP has changed. This determination process is performed based on the presence or absence of a known target recognized by the recognition processing unit 51 in the vicinity of the parking space SP, the presence or absence of an unconfirmed target newly recognized by the recognition processing unit 51, and the like. For example, as shown in FIG. 5, when there is an object X that obstructs the vehicle V on the route from the vehicle entrance B to the parking space SP, the vehicle control unit 53 detects that there is no change in the environment near the parking space SP. determine that there was

If there is no change in the environment around the parking space SP, the vehicle control unit 53 stores the route information acquired by the recognition processing unit 51 in the non-volatile memory 50b of the storage unit 50 in step S160. transition to If the route information has already been stored in the non-volatile memory 50b of the storage unit 50 by the learning process so far, the process proceeds to step S180 without updating the information.

On the other hand, if there is a change in the environment around the parking space SP, the vehicle control unit 53 executes update processing for updating the stored information stored in the storage unit 50 in step S170. In this update process, route update for updating information when the travel route changes and target update for updating information when the target changes are performed. Note that the update process may include not only a process of simply changing information, but also a process of adding information different from previous information. For example, if the information acquisition environment (e.g. season, day/night, weather) differs between the previous information and the new information, each information should be stored without overwriting the previous information. desirable. This is because the greater the amount of information, the higher the accuracy of the route information can be expected.

Below, first, the route update will be described with reference to the flowchart of FIG. As shown in FIG. 6, the vehicle control unit 53 determines whether or not the travel route of the vehicle V has changed in step S200. This determination process is performed, for example, by comparing information about the travel route stored in the storage unit 50 before the current learning process and information about the travel route obtained in the current learning process.

The vehicle control unit 53 skips the subsequent processing if the travel route of the vehicle V has not changed, and proceeds to step S210 if the travel route of the vehicle V has changed. In step S210, the vehicle control unit 53 identifies the difference between the stored information stored in the storage unit 50 before the current learning process and the route information obtained in the current learning process, and converts the difference into different information. Extract as When the traveling route of the vehicle V changes, the distance between the vehicle V and the obstacle OB changes, so that it becomes possible to acquire detailed information such as the shape of the obstacle OB that has not been recognized so far. .

Subsequently, the vehicle control unit 53 updates different information in step S220. In this process, for example, it is desirable to update when different information is continuously detected. The vehicle control unit 53 may update the stored information when there is a large difference between the stored information stored in the storage unit 50 and the route information obtained in the current learning process.

Subsequently, in step S230, the vehicle control unit 53 determines whether or not the update of information has been completed for all sections of the travel route. The vehicle control unit 53 returns to the process of step S210 if the update of the information is not completed, and exits this process if the update of the information is completed.

Next, target updating will be described with reference to the flowchart in FIG. As shown in FIG. 7, the vehicle control unit 53 determines whether or not the target recognized by the recognition processing unit 51 has changed in step S300. This determination process is performed, for example, by comparing the target-related information traveling route stored in the storage unit 50 before the current learning process with the target-related information obtained by the current learning process.

At step S310, the vehicle control unit 53 determines whether or not a known target has not been detected. That is, the vehicle control unit 53 determines whether or not the target detected so far has become undetected.

As a result, if a known target is not detected, there is a high possibility that the object corresponding to the known target has been moved elsewhere by residents. Therefore, when the known target is not detected, the vehicle control unit 53 cancels the registration of the known target in step S320. Specifically, the vehicle control unit 53 deletes the information about the known target from the storage unit 50, and then proceeds to step S330.

On the other hand, if a known target is detected, the vehicle control unit 53 determines in step S340 whether or not a new target has been detected continuously multiple times. That is, the vehicle control unit 53 determines whether or not an unconfirmed target that has not been detected so far has been detected continuously a plurality of times.

As a result, if an unidentified target is detected multiple times in a row, there is a high possibility that an object related to the unidentified target was installed by a resident. Therefore, when an unconfirmed target is detected continuously for a plurality of times, the vehicle control unit 53 newly registers the unconfirmed target in step S350. Specifically, the vehicle control unit 53 stores the information about the unconfirmed target in the storage unit 50 as stored information, and proceeds to step S330.

Here, if the information is updated when the unconfirmed target is a dynamic target, it may adversely affect driving assistance. For this reason, it is desirable that the information is updated when an unidentified target is continuously detected a plurality of times.

At step S330, the vehicle control unit 53 determines whether or not the information has been updated for all sections of the travel route. The vehicle control unit 53 returns to the process of step S310 if the update of the information is not completed, and exits this process if the update of the information is completed.

When the update process is completed as described above, the vehicle control unit 53 proceeds to step S180 in FIG. The vehicle control unit 53 informs the driver via the HMI 45 that the storage of various information is completed, and exits the learning process. In this process, it is preferable that the driver is notified of the storage-related information regarding the storage of the route information in the storage unit 50 by the route storage unit 54 via the HMI 45, which is a notification device. This storage-related information preferably includes at least one of information regarding the start of storing the travel route in the storage unit 50 and information regarding the end of storing the travel route in the storage unit 50 . For example, the vehicle control unit 53 uses the learning start position STP and the learning stop position SPP as memory-related information, and superimposes a mark or frame indicating each position on the camera image displayed on the display of the vehicle V to notify the driver. do. Further, the vehicle control unit 53 may notify the travel route of the vehicle V and the surrounding conditions on the travel route during the learning process. The learning process from step S100 to step S180 is performed by the route storage unit 54 of the vehicle control unit 53. FIG.

Next, the follow-up control process for automatically moving the vehicle V to the parking space SP along the target route will be described with reference to the flowchart shown in FIG. The follow-up control process shown in FIG. 8 is executed by the vehicle control unit 53 in each predetermined control cycle under the condition that the learning process has been performed at least once. Each processing shown in this flowchart is implemented by each functional unit of the driving support device 5 . In addition, each step for realizing this processing can also be grasped as each step for realizing the driving support method.

As shown in FIG. 8, the vehicle control unit 53 starts recognition processing in step S400. In this recognition process, sensing information from the periphery monitoring sensor 3 is input, and the above-described scene recognition, three-dimensional object recognition, and free space recognition are started based on the input sensing information.

Subsequently, in step S410, the vehicle control unit 53 determines whether or not the current position of the vehicle V is near the support start position specified by the driver. Specifically, the vehicle control unit 53 uses sensing information from the surroundings monitoring sensor 3, the map database 35, and the GPS 36 to determine whether or not the current position of the vehicle V is near the support start position. The support start position is set, for example, near the vehicle entrance/exit B that is the boundary between the public road OL and the parking lot PL.

The vehicle control unit 53 waits when the current position of the vehicle V is not near the support start position, and proceeds to step S420 when the current position of the vehicle V is near the support start position. When proceeding to step S420, the vehicle control unit 53 proposes driving assistance for the parking operation to the driver via the HMI 45. FIG. Then, in step S430, the vehicle control unit 53 determines whether or not there is an instruction from the driver via the HMI 45 to perform driving assistance for the parking operation.

The vehicle control unit 53 executes target route generation processing in step S440 when there is an instruction from the driver to perform driving assistance for the parking operation. The process of step S440 is performed by the route generation unit 55 of the vehicle control unit 53. FIG. The vehicle control unit 53 generates a target route that the vehicle V should follow when the vehicle V is parked based on route information including the travel route of the vehicle V during the learning process.

Subsequently, in S450, the vehicle control unit 53 estimates the current position of the vehicle V based on the sensing information stored in the storage unit 50 and the sensing information sequentially acquired by the periphery monitoring sensor 3 during driving assistance.

Subsequently, in step S460, the vehicle control unit 53 automatically moves the vehicle V to the parking space SP by performing vehicle motion control such as acceleration/deceleration control and steering control of the vehicle V. Specifically, the vehicle control unit 53 outputs control signals to various ECUs 4 so that the vehicle V reaches the parking space SP along the target route. After that, in step S470, the vehicle control unit 53 determines whether or not the vehicle V has reached the parking space SP, which is the support end position. The vehicle control unit 53 returns to the process of step S450 when the vehicle V has not reached the parking space SP, and proceeds to step S480 when the vehicle V reaches the parking space SP. The processing of steps S450, S460, and S470 is performed by the follow-up control section 57 of the vehicle control section 53. FIG.

On the other hand, when there is no instruction from the driver to perform driving assistance for the parking operation, the vehicle control unit 53 skips the processes from steps S440 to S470 and proceeds to the process of step S480. In step S480, the vehicle control unit 53 determines whether or not the environment around the parking space SP has changed. This determination process is performed based on the presence or absence of a known target recognized by the recognition processing unit 51 in the vicinity of the parking space SP, the presence or absence of an unconfirmed target newly recognized by the recognition processing unit 51, and the like. This determination process is basically the same as the process of step S150, but may be partially different.

If there is no change in the environment around the parking space SP, the vehicle control unit 53 skips the subsequent processes and exits this process. That is, the vehicle control unit 53 ends the follow-up control process without updating the route information.

On the other hand, if there is a change in the environment around the parking space SP, the vehicle control unit 53 executes update processing for updating the stored information stored in the storage unit 50 in step S490, and ends this processing. In this update process, route update for updating information when the travel route changes and target update for updating information when the target changes are performed. This update process is basically the same as the process in step S170, but may be partially different. The processing of steps S480 and S490 is performed by the route storage unit 54 of the vehicle control unit 53. FIG. The processes of steps S480 and S490 are performed not only when the tracking control unit 57 automatically causes the vehicle V to leave or enter the garage, but also when the driver performs the leaving or entering operation.

The driving support device 5 described above acquires route information including the travel route of the vehicle V and the surrounding conditions of the vehicle V on the travel route, which are successively acquired by the recognition processing unit 51 when the driver performs the leaving operation or the entering operation. A route storage unit 54 is provided for storing information in the storage unit 50 as storage information. The driving assistance device 5 includes a route generation unit 55 that generates a target route that the vehicle V should follow when the vehicle V leaves or enters the vehicle based on the information stored in the storage unit 50 . If the route information newly acquired by the recognition processing unit 51 after the stored information is stored in the storage unit 50 includes information different from the stored information, the route storage unit 54 updates the stored information with the different information. According to this, even if the environment of the parking space SP changes after the stored information is stored in the storage unit 50 , the stored information is updated with the route information newly acquired by the recognition processing unit 51 . Therefore, it is possible to improve robustness against environmental changes in the vicinity of the parking space SP. This environmental change includes not only the presence or absence of other vehicles, bicycles, etc. parked on the road, but also, for example, seasonal changes in color.

Further, according to the driving support device 5 of this embodiment, the following effects can be obtained.

(1) The route storage unit 54 stores the stored information with different information when the travel route when the driver performs the leaving operation or the parking operation after storing the stored information in the storage unit 50 is different from the one stored in the stored information. Update information. The travel route of the vehicle V when the driver leaves or enters the garage is not always constant, and is likely to change in some way, so this timing is suitable for updating the stored information. Therefore, it is desirable to update the stored information with the route information newly acquired by the recognition processing unit 51 when the driver performs the leaving operation or the entering operation. Thereby, generation of an appropriate target route can be expected.

(2) The driving support device 5 includes a follow-up control unit 57 that automatically causes the vehicle V to leave or enter based on the current position of the vehicle V and the target route. If the route information newly acquired by the recognition processing unit 51 when the tracking control unit 57 automatically leaves or enters the vehicle V includes information different from the stored information, the route storage unit 54 stores the different information. to update the stored information. When leaving or entering the parking space is automatically performed by the follow-up control unit 57, the vehicle V basically moves along the target route. change can be easily detected. For this reason, it is desirable to update the stored information with the route information newly acquired by the recognition processing unit 51 when the follow-up control unit 57 automatically performs delivery and storage. Thereby, generation of an appropriate target route can be expected.

(3) If the information on known targets included in the stored information is not included in the route information newly acquired by the recognition processing unit 51, the route storage unit 54 stores the information on known targets as stored information. delete from As a result, the target route can be generated based on information that reflects the actual situation in the vicinity of the parking space SP.

(4) In the route storage unit 54, the route information acquired by the recognition processing unit 51 after the storage information is stored in the storage unit 50 includes information on unconfirmed targets that is not included in the stored information, and is consecutively included a plurality of times. If so, add information about the unidentified target to the stored information. As a result, the target route can be generated based on information that reflects the actual situation in the vicinity of the parking space SP. In particular, in order to determine whether an unidentified target is static or dynamic, it is desirable to add it to the stored information when it is recognized continuously multiple times.

(Modification)
For example, as shown in FIG. 9, the driving assistance device 5 may assist the operation of leaving the vehicle V from the parking space SP. Although FIG. 9 illustrates the situation when leaving the garage facing backward, the driving support by the driving support device 5 is not limited to leaving the garage facing backward, but can also be performed when leaving the garage facing forward. can be done.

For example, as shown in FIG. 10, the driving support device 5 stores the route information sequentially acquired by the recognition processing unit 51 as storage information in the storage unit 50 when the driver performs the parking operation of the vehicle V. . The driving support device 5 generates a target route and performs various controls for driving support based on the information stored in the storage unit 50 and the route information sequentially acquired by the recognition processing unit 51 during driving support. . Even in such a modified example, if the driving assistance device 5 performs update processing at one of the timing of learning processing and driving assistance, it is possible to improve robustness against environmental changes in the vicinity of the parking space SP. can be planned.
Further, the traveling direction of the vehicle V does not have to be the same during the learning process and during the driving assistance, and may be different. For example, a learning process may be performed at the time of forward-facing warehousing, and based on the learning result, driving assistance for backward-facing warehousing may be performed. Similarly, learning processing may be performed at the time of backward parking, and based on the learning result, driving assistance for front parking may be performed. These are the same when driving assistance is performed at the time of leaving the garage.

(Other embodiments)
Although representative embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and can be modified in various ways, for example, as follows.
For example, although the driving assistance device 5 of the above embodiment is connected to the map database 35 and GPS 36, this is not essential. The driving support device 5 does not have to be connected to the map database 35 and the GPS 36, for example. The landmark/marker recognition unit 51ac of the driving assistance device 5 is not essential and may be omitted. In addition, although it is desirable to notify the driver of the learning start position STP, the learning stop position SPP, and the update of the route, the notification is not limited to this. may

In the above-described embodiment, the detailed configuration of the driving assistance device 5, the detailed content of the learning process, and the detailed content of the follow-up control process have been described, but they are not limited to these, and some of them may be different. . For example, the update process executed during the learning process or the follow-up control process is configured to perform either one of route update and target update, or to perform update different from route update and target update. It may be. The learning process may be performed each time the learning start position is reached, or may be performed once every multiple times.

In the above-described embodiment, the updating process is executed when the driver performs the leaving operation or the entering operation, but the driving support device 5 is not limited to this. For example, the driving assistance device 5 may not execute the update process when the driver performs a parking exit operation or a parking operation.

In the above-described embodiment, the follow-up control unit 57 executes the updating process when the vehicle V automatically leaves or enters the garage, but the driving assistance device 5 is not limited to this. For example, the driving assistance device 5 may be configured not to execute the update process when the vehicle V automatically leaves or enters the garage by the follow-up control unit 57 .
Further, in the above-described embodiment, when the current position of the vehicle V is near the support start position specified by the driver, the follow-up control process is executed to automatically move the vehicle V to the parking space SP. However, the follow-up control process is not limited to this. The follow-up control process may be started, for example, when the current position of the vehicle V is near the route traveled by the vehicle V during the learning process.

In the above-described embodiment, when the information on the known target included in the stored information is not included in the route information newly acquired by the recognition processing unit 51, the information on the known target is deleted from the stored information. , the driving support device 5 is not limited to this. For example, when the information on the known target included in the stored information is not included in the route information newly acquired by the recognition processing unit 51, the driving support device 5 stores the information on the known target. may be deleted from the stored information.

In the above-described embodiment, when information about an unconfirmed target that is not included in the stored information is included in the route information newly acquired by the recognition processing unit 51 consecutively a plurality of times, the information about the unconfirmed target is added to the stored information. , but the driving support device 5 is not limited to this. For example, if information about an unconfirmed target that is not in the stored information is included in the route information newly acquired by the recognition processing unit 51 even once, the driving support device 5 stores the information about the unconfirmed target as the stored information. may be added to

In the above-described embodiment, an example in which the driving assistance device 5 of the present disclosure is applied to driving assistance in a narrow parking lot PL in front of one's home has been described. It can also be applied to driving assistance in other lands.

In the above-described embodiment, the automatic driving control including the driving support control by the automatic driving system 1 was described, but regardless of whether the driving support is automatic driving or not, the own vehicle is moved and parked at the support end position. It suffices if it includes control for generating a target route for the purpose. For example, simply displaying the target route on the display and using it as an index when the driver enters or exits the parking lot by his or her own driving is also included in driving support, and route tracking control automatically moves the vehicle according to the target route. not required to be done.

It goes without saying that, in the above-described embodiments, the elements that make up the embodiments are not necessarily essential unless explicitly stated as essential or clearly considered essential in principle.

In the above-described embodiments, when numerical values such as the number, numerical value, amount, range, etc. of the constituent elements of the embodiment are mentioned, when it is explicitly stated that they are essential, and in principle they are clearly limited to a specific number It is not limited to that particular number, unless otherwise specified.

In the above-described embodiments, when referring to the shape, positional relationship, etc. of components, etc., the shape, positional relationship, etc., unless otherwise specified or limited in principle to a specific shape, positional relationship, etc. etc. is not limited.

The controller and techniques of the present disclosure are implemented on a dedicated computer provided by configuring a processor and memory programmed to perform one or more functions embodied by the computer program. good too. The controller and techniques of the present disclosure may be implemented in a dedicated computer provided by configuring the processor with one or more dedicated hardware logic circuits. The control unit and method of the present disclosure is a combination of a processor and memory programmed to perform one or more functions and a processor configured by one or more hardware logic circuits. It may be implemented on one or more dedicated computers. The computer program may also be stored as computer-executable instructions on a computer-readable non-transitional tangible recording medium.

[Features of the present disclosure]
The present disclosure discloses the following technical features.

[Disclosure 1]
A driving support device,
A travel route of the vehicle and the travel route sequentially acquired by a recognition processing unit (51) when a driver performs an operation of leaving the vehicle (V) from the parking space (SP) or entering the parking space. a route storage unit (54) for storing route information including the surrounding situation of the vehicle in the storage unit (50) as storage information;
a route generation unit (55) that generates a target route that the vehicle should follow when the vehicle leaves or enters the vehicle based on the stored information stored in the storage unit;
When the route information newly acquired by the recognition processing unit after the stored information is stored in the storage unit includes information different from the stored information, the route storage unit stores the stored information according to the different information. Driving support equipment to be updated.

[Disclosure 2]
The route storage unit stores the stored information in the storage unit, and when the driving route when the driver performs the leaving operation or the entering operation is different from the one stored in the stored information, the The driving assistance device according to Disclosure 1, wherein the stored information is updated with different information.

[Disclosure 3]
A follow-up control unit (57) for automatically exiting or entering the vehicle based on the current position of the vehicle and the target route,
If the route information newly acquired by the recognition processing unit when the vehicle is automatically parked or parked by the tracking control unit includes information different from the stored information, 3. The driving support device according to Disclosure 1 or 2, wherein the stored information is updated with the different information.

[Disclosure 4]
The route storage unit stores the information on the known target in the stored information when the information on the known target included in the stored information is not included in the route information newly acquired by the recognition processing unit. 4. The driving assistance device according to any one of Disclosures 1 to 3, which is deleted from .

[Disclosure 5]
When the route information newly acquired by the recognition processing unit contains information on an unconfirmed target that is not included in the stored information, the route storage unit stores the information on the unconfirmed target. to the stored information.

[Disclosure 6]
A driving assistance method comprising:
A travel route of the vehicle and the travel route sequentially acquired by a recognition processing unit (51) when a driver performs an operation of leaving the vehicle (V) from the parking space (SP) or entering the parking space. storing route information including the surrounding situation of the vehicle in a storage unit (50) as storage information;
generating a target route that the vehicle should follow when leaving or entering the vehicle based on the stored information stored in the storage unit;
When the route information newly acquired by the recognition processing unit after storing the stored information in the storage unit includes information different from the stored information, the driving support method updates the stored information with the different information. .

Claims

A driving support device,
A travel route of the vehicle and the travel route sequentially acquired by a recognition processing unit (51) when a driver performs an operation of leaving the vehicle (V) from the parking space (SP) or entering the parking space. a route storage unit (54) for storing route information including the surrounding situation of the vehicle in the storage unit (50) as storage information;
a route generation unit (55) that generates a target route that the vehicle should follow when the vehicle leaves or enters the vehicle based on the stored information stored in the storage unit;
When the route information newly acquired by the recognition processing unit after the stored information is stored in the storage unit includes information different from the stored information, the route storage unit stores the stored information according to the different information. Driving support equipment to be updated.
The route storage unit stores the stored information in the storage unit, and when the driving route when the driver performs the leaving operation or the entering operation is different from the one stored in the stored information, the 2. The driving assistance device according to claim 1, wherein said stored information is updated with different information.
A follow-up control unit (57) for automatically exiting or entering the vehicle based on the current position of the vehicle and the target route,
If the route information newly acquired by the recognition processing unit when the vehicle is automatically parked or parked by the tracking control unit includes information different from the stored information, The driving assistance device according to claim 1 or 2, wherein said stored information is updated with said different information.
The route storage unit stores the information on the known target in the stored information when the information on the known target included in the stored information is not included in the route information newly acquired by the recognition processing unit. 3. The driving support device according to claim 1 or 2, deleted from.
When the route information newly acquired by the recognition processing unit contains information on an unconfirmed target that is not included in the stored information, the route storage unit stores the information on the unconfirmed target. to the stored information.
A driving assistance method comprising:
A travel route of the vehicle and the travel route sequentially acquired by a recognition processing unit (51) when a driver performs an operation of leaving the vehicle (V) from the parking space (SP) or entering the parking space. storing route information including the surrounding situation of the vehicle in a storage unit (50) as storage information;
generating a target route that the vehicle should follow when leaving or entering the vehicle based on the stored information stored in the storage unit;
When the route information newly acquired by the recognition processing unit after storing the stored information in the storage unit includes information different from the stored information, the driving support method updates the stored information with the different information. .