WO2019123570A1

WO2019123570A1 - Parking position determination device and parking position determination program

Info

Publication number: WO2019123570A1
Application number: PCT/JP2017/045726
Authority: WO
Inventors: 村下　君孝
Original assignee: 富士通株式会社
Priority date: 2017-12-20
Filing date: 2017-12-20
Publication date: 2019-06-27

Abstract

Disclosed is a parking position determination device including: a storage unit that stores, for each attribute of an object of interest and in association with the attribute, the distance information representing the distance from a vehicle reference to the object of interest in the direction of the vehicle length; and a processing unit that acquires from the storage unit the distance information associated with the attribute of the object of interest which, according to a captured camera image of the vehicle surroundings, occupies a space in the direction of the vehicle length and uses the acquired distance information to determine a target stop position, along the direction of the vehicle length, at the completion of parking .

Description

Parking position determination device and parking position determination program

The present disclosure relates to a parking position determination device and a parking position determination program.

2. Description of the Related Art Conventionally, there is known a technique of detecting a parking frame or the like with a camera to determine a target stop position (a target vehicle position related to the longitudinal direction of the vehicle at the completion of parking).

JP, 2009-202610, A JP 2002-154396 A JP, 2010-164349, A

However, in the above-mentioned prior art, it is difficult to determine an appropriate target stop position according to the attribute of the object that may exist in the space in the vehicle traveling direction. For example, in the space in the vehicle traveling direction, there may be an object having different attributes such as a parking frame or a ring, and an appropriate target stop position may be different depending on the attribute of the object.

Therefore, in one aspect, the present invention aims to determine an appropriate target stop position according to the attribute of an object that may be present in the space in the vehicle traveling direction.

According to one aspect of the present disclosure, a storage unit that stores, for each attribute of an object, distance information representing a distance in the vehicle front-rear direction from the vehicle reference to the object in association with the attribute;
The distance information associated with the attribute of the object present in the space in the traveling direction of the vehicle is acquired from the storage unit based on the image of the camera obtained by imaging the periphery of the vehicle, and parking is completed using the acquired distance information A parking position determination device is provided that includes a processing unit that determines a target stop position related to the front-rear direction of the vehicle.

According to the present disclosure, it is possible to determine an appropriate target stop position according to the attribute of an object that may exist in the space in the vehicle traveling direction.

FIG. 1 schematically shows an overall configuration of a vehicle system 1 according to an embodiment. It is a figure which shows an example of the functional block of the sensing control apparatus. It is explanatory drawing of the derivation | leading-out method of the distance of a target object. It is explanatory drawing of the derivation | leading-out method of the distance of a target object. It is a figure showing an example of a predetermined object (and the 2nd obstacle). It is a figure which shows another example of a predetermined | prescribed object (and 2nd obstruction). It is an example of a definition table (distance information). It is explanatory drawing of the distance information matched with a parking frame. It is explanatory drawing of the distance information matched with a loop. It is explanatory drawing of the determination method of the target stop position based on a parking frame. It is explanatory drawing of the determination method of the target stop position based on looping. It is explanatory drawing of the determination method of the target stop position based on a 1st obstacle. It is explanatory drawing of the determination method of the target stop position based on a 2nd obstacle. It is explanatory drawing of the determination method of the target stop position when the target object of a several different attribute is detected. It is a schematic flowchart which shows the operation example of the sensing control apparatus 10 by a present Example. It is a schematic flowchart which shows an example of the target stop position calculation process which concerns on a target object. It is a schematic flowchart which shows an example of the target stop position calculation process which concerns on a 2nd obstruction.

Hereinafter, each example will be described in detail with reference to the attached drawings.

FIG. 1 is a diagram schematically showing an overall configuration of a vehicle system 1 according to an embodiment. An example of a hardware configuration is schematically shown in FIG. 1 for a part of the configuration.

The vehicle system 1 is mounted on a vehicle. The vehicle system 1 includes a sensing control device 10 (an example of a parking position determination device), a vehicle control device 12, a camera 14, a steering wheel control device 16, a brake / accelerator control device 18, and a monitor 19.

The sensing control device 10 is, for example, in the form of an electronic control unit (ECU), and has a hardware configuration as illustrated in FIG. 1, for example. In the example illustrated in FIG. 1, the sensing control device 10 includes a central processing unit (CPU) 110 (an example of a processing unit), a random access memory (RAM) 111, and a read only memory (ROM) 112 (storage An example of a part) etc.

The vehicle control device 12, the camera 14, and the monitor 19 are connected to the sensing control device 10. For example, the vehicle control device 12 is connected to the sensing control device 10 via a CAN (controller area network), and the sensing control device 10 is monitored via an AVC-LAN (Audio-Visual Communication Local Area Network). 19 is connected.

The sensing control device 10 performs a parking position determination process of determining a target parking position and a target parking direction based on an image of the camera 14. The parking position determination process will be described later. The sensing control device 10 provides the result of the parking position determination process to the steering wheel control device 16 and the brake / accelerator control device 18.

The vehicle control device 12 is, for example, in the form of an ECU like the sensing control device 10, and has a hardware configuration as shown in FIG. 1, for example. A steering wheel control device 16 and a brake / accelerator control device 18 are connected to the vehicle control device 12.

The camera 14 is, for example, a monocular camera, and is attached to the rear of the vehicle. In this embodiment, as an example, in a state where the camera 14 is attached, the optical axis of the camera 14 is in a fixed state (that is, the optical axis of the camera 14 is not variable). The camera 14 images the rear of the vehicle. The camera 14 gives the captured image to the sensing control device 10. In a modification, the camera 14 may be another type of camera such as a stereo camera. Also, the vehicle system 1 may further include another camera attached to the vehicle side or the front of the vehicle.

The steering wheel control device 16 rotates the turning wheels of the vehicle until the vehicle reaches the target parking position so that the vehicle becomes the target parking direction at the target parking position according to the result of the parking position determination process from the vehicle control device 12 Control the rudder direction (ie, the direction of travel of the vehicle). For example, the steering wheel control device 16 controls the turning direction of the turning wheels via a power steering device (not shown).

The brake / accelerator control device 18 causes the vehicle to reach the target parking position so that the vehicle moves to the target parking position and stops at the target parking position according to the result of the parking position determination process from the vehicle control device 12 Control the braking and driving force of the vehicle up to For example, the brake / accelerator control device 18 generates a braking force via a brake actuator (not shown) or generates a driving force via a driving force generator (for example, an engine, a motor, etc.) .

The functions of the vehicle control device 12, the steering wheel control device 16, and the brake / accelerator control device 18 may be integrally realized by one control device or partially realized by another control device. It is also good. For example, some or all of the functions of the vehicle control device 12 may be realized by the steering wheel control device 16 and / or the brake and accelerator control device 18. In addition, the vehicle control device 12 may be realized by a combination of a plurality of control devices.

The monitor 19 is provided in the vehicle compartment. The monitor 19 may be a liquid crystal display or the like. The monitor 19 may be realized by a display of a portable terminal that the user brings into the vehicle.

FIG. 2 is a diagram showing an example of a functional block of the sensing control device 10. As shown in FIG.

The sensing control device 10 includes an image dividing unit 101, a first object detecting unit 102, a first distance measuring unit 103, a predetermined object detecting unit 104, a second object detecting unit 105, and a second distance measuring unit 106. , And a target parking position calculation unit 107. Each of the units 101 to 107 can be realized, for example, by the CPU 110 executing a program in the ROM 112.

Further, the sensing control device 10 includes a definition table holding unit 108. The definition table storage unit 108 can be realized by the ROM 112. In a modification, the definition table storage unit 108 may be realized by a flash memory, or an auxiliary storage device such as a hard disk drive (HDD) or a solid state drive (SSD).

Based on steering angle information from a steering angle sensor (not shown), the image dividing unit 101 generates an image from the camera 14 as an image area (hereinafter referred to as “first image area”) relating to a space within the vehicle travel range. And an image area (hereinafter referred to as a "second image area") related to a space outside the vehicle travel range. The image area related to the space within the vehicle travel range is an image area related to the space where the vehicle is scheduled to pass based on the traveling direction of the vehicle, and is set in a mode to which a predetermined margin is added. A space in which the vehicle is scheduled to pass based on the traveling direction of the vehicle can be derived based on the steering angle information.

In the modification, the image dividing unit 101 may divide the image from the camera 14 into an image area related to the parking space and an image area related to the space outside the parking space. The parking space is a space where a vehicle at the time of parking completion is to be present, and is set in a mode to which a predetermined margin is added. The parking space may be set based on an image from the camera 14 or may be set based on another sensor (for example, an ultrasonic sensor).

The space in the vehicle travel range and the parking space are examples of the space in the vehicle traveling direction.

The first object detection unit 102 detects (recognizes) an object to be detected (hereinafter, referred to as an “object”) in the first image region, and detects (recognizes) an attribute (type) of the object. .

In this embodiment, as an example, the attribute of the object includes a parking frame having a frame (for example, a white line painted on a road surface), an obstacle, and a loop. An obstacle is an object that a vehicle should not hit or that is not desirable to hit, such as other vehicles, people, walls, fences, bicycles, two-wheeled vehicles, fixed objects with relatively high height, moving objects, etc. Therefore, for example, an object (a character, a sign, etc.) having no thickness on the road surface other than the parking frame, a bounce board of the time-type parking lot, and the like do not correspond to the obstacle. Note that these objects are not included in the object, but may be detected for other applications.

The first object detection unit 102 detects the presence or absence of an object, and the attribute of the object if it exists, using, for example, HoG (Histogram of oriented gradients) or deep learning. Hereinafter, the obstacle among the objects detected by the first object detection unit 102 is also referred to as a “first obstacle”.

The first distance measuring unit 103 measures the positional relationship of the vehicle with respect to the object. For example, the first distance measuring unit 103 derives the coordinate value of the object in a plane coordinate system in top view with the center of the rear wheel axle of the vehicle as the origin. For example, SFM (Structure from Motion) or moving stereo can be used to derive the coordinate values of the object.

For example, when using a monocular camera, the first distance measuring unit 103 first calculates the position (coordinates of the image) from the image center of the object (step S300), as schematically shown in FIGS. 3A and 3B, for example. Then, based on the lens distortion, an angle α (see FIG. 3B) from the optical axis to the contact position of the object is derived (step S301). Lens distortion is an internal parameter and is determined by the lens design of the camera 14. The first distance measuring unit 103 uses, for example, a table (not shown) that defines the relationship between the distance (the number of pixels) from the image center and the angle from the camera optical axis to determine the angle from the optical axis. It may be derived. Next, the first distance measuring unit 103 derives the distance d of the contact position of the object based on the external parameter. The external parameters are the mounting position (height) of the camera 14, the mounting angle (yaw, pitch, roll) of the camera 14, the physical mounting position of the camera 14, and are determined by the camera mounting design.

The predetermined object detection unit 104 detects a predetermined object in the second image area. The predetermined object is in the form of a column extending in the vertical direction, and at least two or more objects are apart from each other in the direction crossing the traveling direction of the vehicle. In this embodiment, as an example, the predetermined object includes a pole 300 for passing a chain or a rope as shown in FIG. 4A, and a support 301 for supporting a net (fence) as shown in FIG. 4B. .

The pole 300 has a feature that the height is relatively low, pairs of two each form a pair, and a line segment connecting the pole 300 is perpendicular to the traveling direction of the vehicle in the parking frame. The predetermined object detection unit 104 detects the pole 300 using such features. The columns 301 are relatively high (e.g., higher than the average vehicle height), and are paired in pairs, and the line connecting the columns 301 is perpendicular to the vehicle traveling direction in the parking frame There is a feature. The predetermined object detection unit 104 detects the pole 300 using such features. The predetermined object detection unit 104 does not need to detect the pole 300 and the support column 301 in a distinguishable manner.

Here, when the pole 300 and the support column 301 exist in the space in the vehicle traveling direction, the pole 300 and the support column 301 are highly likely to be detected as a first obstacle by the first object detection unit 102. When the pole 300 and the support column 301 do not exist in the space in the vehicle traveling direction, the possibility that the predetermined object detection unit 104 detects a predetermined object is high.

The second object detection unit 105 is based on the detection result (detection result) of the predetermined object by the predetermined object detection unit 104, and is an obstacle that exists in the vehicle traveling direction and can not be recognized based on the image of the camera Therefore, it is presumed to detect “a second obstacle”). An obstacle that can not be recognized based on the image of the camera 14 means an obstacle that can not be detected by the first object detection unit 102 or an obstacle that can not be detected with the required accuracy. Thus, the second obstacle is a new object different from the first obstacle.

In the present embodiment, as an example, the second obstacle includes a rope or chain 320 (see FIG. 4A) provided between the poles 300 and a net 322 (see FIG. 4B) provided between the support columns 301. These ropes and chains 320, etc. are objects significantly thinner than the pole 300 and the support column 301, and can not be detected by the first object detection unit 102 with high accuracy. This is because the area occupied by the rope in the image is less than 3%, and it is impossible to distinguish between the rope, the road surface pattern, and the road surface crack. The same applies to the chain 320 and the net 322.

The second object detection unit 105 detects that there is a second obstacle on a line connecting two or more predetermined objects in top view, and detects the second obstacle. Therefore, detection of the second obstacle by the second object detection unit 105 is based on estimation only.

The second ranging unit 106 measures the positional relationship of the vehicle with respect to the second obstacle. For example, the second distance measuring unit 106 derives the coordinate value of the second obstacle in the plane coordinate system in top view with the center of the rear wheel axle of the vehicle as the origin. The coordinate values of the second obstacle may be expressed, for example, as coordinates on a line connecting two or more predetermined objects.

The target parking position calculation unit 107 determines a target parking position and a target stopping direction (target parking direction) based on the coordinate value of the object and the coordinate value of the second obstacle as the parking position determination processing. In the following, as an example, the target parking position represents a target position to which the rear wheel axle center of the vehicle should reach. In the following, as an example, the target parking position is derived in a plane coordinate system in top view with the center of the rear wheel axle of the vehicle as the origin. In this case, parking is completed when the target parking position reaches the origin.

The target parking position includes a target stop position and a target lateral position. The target stop position is a target value of a position relating to the longitudinal direction of the vehicle at the time of parking completion, and the target lateral position is a target value of a position relating to the lateral direction of the vehicle at the completion of parking. The target stop direction is a target value of the direction of the vehicle when parking is completed.

The method of determining the target stop position will be described in detail later. Although the method of determining the target lateral position is arbitrary, the target lateral position may be set, for example, at the center between the parking frames on both sides in the lateral direction or at the center in the lateral direction of the parking space. Although the method of determining the target stopping direction is arbitrary, the target stopping direction may be set, for example, parallel to the parking frame on both sides in the lateral direction, or may be set parallel to the longitudinal direction of the parking space.

The definition table storage unit 108 stores distance information in association with an attribute for each attribute of the object. FIG. 5A is an example of the definition table held by the definition table holding unit 108. In FIG. 5A, classification symbols a-1, a-2, a-3, b and c are given for each attribute of the object, and the vehicle standard and the distance from the object are defined for each attribute (definition ). For example, the classification symbol a-1 corresponds to the attribute of the parking frame, the classification symbol a-2 corresponds to the attribute of the first obstacle, and the classification symbol a-3 corresponds to the attribute of the loop. The classification symbol c corresponds to the attribute of the second obstacle. Although the first obstacle and the second obstacle are treated as different attributes in FIG. 5A, they may be treated as the same attribute.

In the definition table of FIG. 5A, the vehicle reference is the rear end of the vehicle or the rear wheel axle of the vehicle, and the distance information is the distance from the rear end of the vehicle or the rear wheel axle of the vehicle to the object in the longitudinal direction of the vehicle Represents the distance to be realized when the vehicle is stopped.

Specifically, in the distance information associated with the parking frame, the frame line in the direction crossing the traveling direction of the vehicle and the rear end of the vehicle (an example of the vehicle end in the front side in the traveling direction of the vehicle) substantially match Defined in In FIG. 5A, as the distance information associated with the parking frame, for example, as schematically shown in FIG. 5B, the distance between the white line (an example of a frame line) and the rear end of the vehicle is “0 [cm]”. In FIG. 5B, the vehicle is shown in a state in which the distance from the rear end of the vehicle to the white line is 0 cm. However, in a modification, the distance may be set to be slightly larger than “0 [cm]”.

The distance information associated with the first obstacle is defined such that the vehicle does not hit the first obstacle. That is, the distance information associated with the first obstacle is defined so that the rear end of the vehicle is separated from the obstacle by a predetermined distance or more. In FIG. 5A, the distance information associated with the first obstacle is such that the distance from the rear end of the vehicle is “10 [cm] (an example of a predetermined distance)”. The distance information associated with the first obstacle is not limited to "10 cm", and may be set according to a necessary margin or the like, or may be set by the user.

The distance information associated with the detent is defined such that the rear wheel of the vehicle (an example of the front wheel in the traveling direction of the vehicle) is in a predetermined positional relationship such that it contacts or nearly touches the detent. In FIG. 5A, as an example, as shown schematically in FIG. 5C, for example, the distance information from the rear wheel shaft is “30 [cm]” as the distance information associated with the loop. The distance information associated with the clasping is not limited to "30 [cm]", and may be set according to the tire diameter and the like.

In the modification, the distance information associated with the ringing may be defined by the distance from the rear end of the vehicle. This is because the distance between the rear end of the vehicle and the rear wheel shaft (the distance in the longitudinal direction of the vehicle) is constant (design value), and therefore, it can function based on either of them. Similarly, portions other than the rear end of the vehicle and the rear wheel of the vehicle may be used as a reference. The same applies to the distance information associated with the parking frame and the distance information associated with the first obstacle and the second obstacle. Also, from the same point of view, the vehicle reference may be an arbitrary position in a fixed relationship with the rear end of the vehicle.

The distance information associated with the second obstacle is defined such that the vehicle does not hit the second obstacle. That is, the distance information associated with the second obstacle is defined so that the rear end of the vehicle is separated from the obstacle by a predetermined distance or more. In FIG. 5A, the distance information associated with the second obstacle is such that the distance from the rear end of the vehicle is “10 [cm] (an example of a predetermined distance)”. The distance information associated with the second obstacle is not limited to "10 cm", and may be set according to a necessary margin or the like, or may be set by the user.

Next, a method of determining the target stop position will be described with reference to FIGS. In addition, the line (line L1 etc.) demonstrated below is a line orthogonal to a target parking direction by top view. 6 to 9 show a picture of the vehicle when going straight and moving backward, but the present invention is similarly applicable to moving backward with turning.

In the present embodiment, the target parking position calculation unit 107 acquires distance information associated with the attribute of the object present in the parking space from the definition table holding unit 108, and uses the acquired distance information to obtain the target stop position. decide.

FIG. 6 is an explanatory view of a method of determining the target stop position when the parking frame is detected, and is a schematic top view of the parking lot. FIG. 6 shows a vehicle that is going to park in a backward movement to the parking space S6.

In this case, only the parking frame 600 is present as an object in the parking space S6. Therefore, based on the distance information associated with the parking frame, the target stop position is a predetermined distance from the parking frame 600 on the back side of the parking space S6. Only the β is set on the entrance side of the parking space S6. The line L1 indicates a position on the entrance side of the parking space S6 by a predetermined distance β from the parking frame 600 on the back side of the parking space S6. The predetermined distance β is the distance from the rear end of the vehicle to the center of the rear wheel axle of the vehicle (the distance in the longitudinal direction of the vehicle). In this case, the target stop position when the parking frame is detected is determined on the line L1.

FIG. 7 is an explanatory view of a method of determining the target stop position when the clasping is detected, and is a schematic top view of the parking lot. FIG. 7 shows a vehicle which is going to park in a backward movement to the parking space S7.

In this case, since only the clasp 700 exists as an object in the parking space S7, based on the distance information associated with the clasp, the target stop position is "30 [cm]" parked from the clasp 700 It is set to the space S7 entrance side. The line L2 indicates the position of the snap ring 700, and the line L3 indicates a position on the entrance side of the parking space S7 by "30 cm" from the snap ring 700. In this case, the target stop position when the snap ring 700 is detected is determined on the line L3.

FIG. 8 is an explanatory diagram of a method of determining the target stop position when the first obstacle is detected, and is a schematic top view of the parking lot. FIG. 8 shows a vehicle which is going to park in a backward traveling to the parking space S8. In FIG. 8, a first obstacle 800 is schematically shown in the parking space S8.

In this case, the first obstacle 800 is detected (see line L8). The line L9 indicates a position at which the distance from the line L8 is “10 [cm]” based on the distance information associated with the first obstacle. The target stop position is set on the entrance side of the parking space S8 by a predetermined distance β from the line L9. Line L10 indicates a position on the entrance side of parking space S8 by a predetermined distance β from line L9. The predetermined distance β is the distance from the rear end of the vehicle to the center of the rear wheel axle of the vehicle (the distance in the longitudinal direction of the vehicle). In this case, the target stop position when the first obstacle 800 is detected is determined on the line L10.

FIG. 9 is an explanatory diagram of a method of determining the target stop position when the second obstacle is detected, and is a schematic top view of the parking lot. FIG. 9 shows a vehicle which is going to park in a backward traveling to the parking space S9. Two poles 300 are schematically shown in FIG.

In this case, the two poles 300 exist in the space S61 outside the vehicle travel range, and do not exist in the space (and the parking space S9) within the vehicle travel range. Therefore, in this case, the two poles 300 are detected as a predetermined object, and the presence of the second obstacle is estimated between the two poles 300 (see line L4). The line L5 indicates the position at which the distance from the line L4 is "10 cm", based on the distance information associated with the second obstacle. The target stop position is set on the entrance side of the parking space S9 by a predetermined distance β from the line L5. Line L6 indicates a position on the entrance side of parking space S9 by a predetermined distance β from line L5. The predetermined distance β is the distance from the rear end of the vehicle to the center of the rear wheel axle of the vehicle (the distance in the longitudinal direction of the vehicle). In this case, the target stop position when the second obstacle is detected is determined on the line L6.

Thus, according to the present embodiment, an appropriate target stop position can be determined according to the attribute of the object that may exist in the parking space. For example, it is possible to determine a target stop position where the rear end of the vehicle substantially matches the parking frame, or a target stop position where the rear wheel just touches or almost touches the loop. Also, the target stop position can be determined at a position separated by an appropriate distance from the first obstacle.

Further, according to this embodiment, it is possible to detect a second obstacle (for example, a chain, a rope, a net or the like) which can not be recognized by ordinary image recognition, based on the detection result of a predetermined object such as a pole or a support. This makes it possible to determine an appropriate target stop position even for a parking space in which the second obstacle is present. That is, the target stop position can be determined at a position separated by an appropriate distance from the second obstacle.

Next, with reference to FIG. 10, a method of determining the target stop position when objects of a plurality of different attributes are detected will be described.

FIG. 10 is an explanatory view of a method of determining the target stop position when objects of a plurality of different attributes are detected, and is a schematic top view of the parking lot. FIG. 10 shows a vehicle which is going to park in a backward movement to the parking space S10.

When objects of a plurality of different attributes are detected, the target stop position determined as described above is calculated as a candidate for each attribute, and the candidate on the parking space entrance side among the plurality of candidates is the target stop It is determined as a position.

In FIG. 10, the line L20 is a line on which the candidate of the target stop position obtained according to the method of determining the target stop position when the parking frame is detected is located. The line L21 is a line on which candidates for the target stop position obtained according to the method of determining the target stop position when the loop is detected are located. The line L22 is a line on which candidates for the target stop position obtained according to the method for determining the target stop position when the first obstacle is detected are located. The line L23 is a line on which candidates for the target stop position obtained according to the method of determining the target stop position when the second obstacle is detected are located. In this case, the target stop position is determined on the candidate at the parking space entrance side, ie, the line L20.

As a result, when a plurality of targets of different attributes are detected, it is possible to determine the most safe candidate as the target stop position among the plurality of candidates for the target stop position. For example, it is assumed that the line on which the candidate for the target stop position based on the parking frame is located is behind the parking space than the line on which the candidate for the target stop position based on ringing is located. In this case, if the target stop position is determined on the line on which the candidate for the target stop position based on the parking frame is located, the rear wheel will get over the ringing by the time the vehicle reaches the target stop position, which is not preferable. According to the present embodiment, such inconvenience can be prevented.

Although four candidates are calculated in the example of FIG. 10, the same applies to the case where two or three candidates are calculated.

Next, with reference to FIG. 11 or later, an operation example of the sensing control device 10 according to the present embodiment will be described using a flowchart.

FIG. 11 is a schematic flowchart showing an operation example of the sensing control device 10 according to the present embodiment.

In step S1100, the image dividing unit 101 of the sensing control device 10 divides the image from the camera 14 into a first image area and a second image area.

In step S1102, the sensing control device 10 performs a target stop position calculation process related to the object based on the first image area. An example of the target stop position calculation process which concerns on a target object is later mentioned with reference to FIG.

In step S1104, the predetermined object detection unit 104 of the sensing control device 10 performs processing for detecting a predetermined object in the second image area.

In step S1106, the sensing control device 10 performs target stop position calculation processing relating to the second obstacle based on the processing result in step S1104. An example of the target stop position calculation process according to the second obstacle will be described later with reference to FIG.

In step S1108, the target parking position calculation unit 107 of the sensing control device 10 performs a parking position determination process. At this time, when there are a plurality of candidates for the target stop position, the target parking position calculation unit 107 determines the candidate closest to the parking space entrance as the target stop position (see FIG. 10).

FIG. 12 is a schematic flowchart showing an example of the target stop position calculation process (step S1102 in FIG. 11) related to the object.

In step S1200, the first object detection unit 102 performs processing for detecting a parking frame in the first image area, and when a parking frame is detected, the process proceeds to step S1202. If the parking space is not detected, the process proceeds to step S1208.

In step S1202, the first distance measuring unit 103 measures the positional relationship of the vehicle with respect to the parking frame. For example, the first distance measuring unit 103 derives the coordinate value of the parking frame on the back side of the parking space in the plane coordinate system in top view with the center of the rear wheel shaft of the vehicle as the origin.

In step S1204, the target parking position calculation unit 107 acquires, from the definition table storage unit 108, the distance information associated with the parking frame.

In step S1206, the target parking position calculation unit 107 is a candidate for a target stop position based on the parking frame based on the coordinate values of the parking frame on the back side of the parking space obtained in step S1202 and the distance information obtained in step S1204. Calculate The calculation method of the candidate of the target stop position based on a parking frame is as above-mentioned (refer FIG. 6).

In step S1208, the first object detection unit 102 performs a process of detecting the first obstacle in the first image region, and when the first obstacle is detected, the process proceeds to step S1210. If the first obstacle is not detected, the process proceeds to step S1216.

In step S1210, the first distance measuring unit 103 measures the positional relationship of the vehicle with respect to the first obstacle. For example, the first distance measuring unit 103 derives the coordinate value of the first obstacle (for example, the coordinate value on the near side of the parking space) in the plane coordinate system in top view with the center of the rear wheel axle of the vehicle as the origin.

In step S1212, the target parking position calculation unit 107 acquires, from the definition table holding unit 108, distance information associated with the first obstacle.

In step S1214, the target parking position calculation unit 107 selects a candidate for a target stop position based on the first obstacle based on the coordinate value of the first obstacle obtained in step S1210 and the distance information obtained in step S1212. calculate. The calculation method of the candidate of the target stop position based on the first obstacle is as described above (see FIG. 8).

In step S1216, the first object detection unit 102 performs a process of detecting looping in the first image region, and when looping is detected, the process proceeds to step S1218. If the ringing is not detected, the process ends as it is (the process proceeds to step S1104 in FIG. 11).

In step S1218, the first distance measuring unit 103 measures the positional relationship of the vehicle relative to the ring. For example, the first distance measuring unit 103 derives a coordinate value (for example, a coordinate value on the near side of the parking space) of the ringing in a plane coordinate system in top view with the center of the rear wheel axle of the vehicle as an origin.

In step S1220, the target parking position calculation unit 107 acquires, from the definition table storage unit 108, the distance information associated with the loop.

In step S1222, the target parking position calculation unit 107 calculates a candidate for a target stop position based on the looping based on the looping coordinate values obtained in step S1218 and the distance information obtained in step S1220. The calculation method of the candidate of the target stop position based on the clasp is as described above (see FIG. 7).

In addition, in FIG. 12, although the candidate of the target stop position which concerns on each attribute is calculated in order of a parking frame, a 1st obstacle, and ringing, a processing order is arbitrary.

FIG. 13 is a schematic flowchart showing an example of the target stop position calculation process (step S1106 in FIG. 11) related to the second obstacle.

In step S1300, the second object detection unit 105 determines whether the predetermined object detection unit 104 has detected a predetermined object. If the determination result is "YES", the process proceeds to step S1302, and otherwise, the process ends as it is (process proceeds to step S1108 in FIG. 11).

In step S1302, the second object detection unit 105 detects the second obstacle on a line connecting two or more predetermined objects.

In step S1304, the second distance measuring unit 106 measures the positional relationship of the vehicle with respect to the second object. For example, the second distance measuring unit 106 derives the coordinate value of the second object (for example, the coordinate value on the near side of the parking space) in the plane coordinate system in top view with the center of the rear wheel axle of the vehicle as the origin.

In step S1306, the target parking position calculation unit 107 acquires, from the definition table storage unit 108, distance information associated with the second obstacle.

In step S1308, the target parking position calculation unit 107 selects a candidate for a target stop position based on the second obstacle based on the coordinate value of the second obstacle obtained in step S1304 and the distance information obtained in step S1306. calculate. The method of calculating the candidate for the target stop position based on the second obstacle is as described above (see FIG. 9).

As mentioned above, although each Example was explained in full detail, it is not limited to a specific example, A various deformation | transformation and change are possible within the range described in the claim. In addition, it is also possible to combine all or a plurality of the components of the above-described embodiment.

For example, in the embodiment described above, the second obstacle is detected as a preferred example, but the detection of the second obstacle is not essential.

Also, in the embodiment described above, the attribute of the object is three types of parking frame, ringing and obstacles, but four or more types may be set, and of the parking frame, ringing and obstacles There may be two types.

Further, in the above-described embodiment, the definition table (distance information corresponding to the attribute) defined in the definition table storage unit 108 is fixed, but may be updated by learning or the like.

In the above-mentioned embodiment, although parking by retreat is illustrated, it is applicable also to forward parking. In this case, the same function can be realized by using an image obtained by the front camera instead of the camera 14. Further, in this case, the vehicle end on the front side in the vehicle traveling direction is the front end of the vehicle, and the wheel on the front side in the vehicle traveling direction is a front wheel.

Also, in the embodiment described above, the target stop position determined as described above is used for automatic parking, but may be used for semi-automatic parking or parking assistance. For example, in parking assistance in which the driving force, the braking force, and the steering are adjusted by the driver's operation, the target stop position determined as described above may be displayed on the monitor 19. Then, immediately before the vehicle reaches the target stop position, support information for prompting the brake operation may be displayed on the monitor 19 or may be output by voice or the like.

DESCRIPTION OF SYMBOLS 1 vehicle system 10 sensing control device 12 vehicle control device 14 camera 16 steering wheel control device 18 brake and accelerator control device 19 monitor 101 image dividing unit 102 first object detecting unit 103 first distance measuring unit 104 predetermined object detecting unit 105 second object Detection unit 106 Second ranging unit 107 Target parking position calculation unit 108 Definition table holding unit 300 Pole 301 Support column 600 Parking frame 800 First obstacle

Claims

A storage unit that stores, for each attribute of the object, distance information indicating a distance from the vehicle reference to the object in the vehicle front-rear direction in association with the attribute;
The distance information associated with the attribute of the object present in the space in the traveling direction of the vehicle is acquired from the storage unit based on the image of the camera obtained by imaging the periphery of the vehicle, and parking is completed using the acquired distance information And a processing unit that determines a target stop position related to the front-rear direction of the vehicle.
The parking position determination device according to claim 1, wherein the vehicle reference is a rear end of a vehicle or a rear wheel axle.
The attributes of the object include a parking frame with a border and a ring connection,
The distance information associated with the parking frame is defined such that the frame line in the direction intersecting the vehicle traveling direction substantially matches the front end of the vehicle in the vehicle traveling direction.
The parking position determination device according to claim 1, wherein the distance information associated with the ring connection is defined such that a rear wheel of the vehicle and the ring connection have a predetermined positional relationship.
The attributes of the object include a parking frame with a border, a ring and an obstacle,
The distance information associated with the parking frame is defined such that the frame line in the direction intersecting the vehicle traveling direction substantially matches the front end of the vehicle in the vehicle traveling direction.
The distance information associated with the ringing is defined such that a front wheel in the vehicle traveling direction and the ringing have a predetermined positional relationship.
The parking position determination according to claim 1, wherein the distance information associated with the obstacle is defined such that a vehicle end on the front side in the vehicle traveling direction is in a relationship of being separated from the obstacle by a predetermined distance or more. apparatus.
The obstacle is present in a space in the vehicle traveling direction and can be recognized based on an image of the camera, and in a space in the vehicle traveling direction and can not be recognized based on an image of the camera The parking position determination apparatus according to claim 4, further comprising:
The parking position determination device according to claim 5, wherein the processing unit detects the second obstacle based on a detection result of a predetermined object that does not exist in the space in the vehicle traveling direction.
The predetermined object is in the form of a column extending in the vertical direction, and at least two or more of the predetermined objects are separated in a direction intersecting the traveling direction of the vehicle,
The parking position determination device according to claim 6, wherein the second obstacle is in the form of at least one of a rope, a chain, and a net provided between two or more of the predetermined objects.
The processing unit calculates a candidate for a target stop position for each attribute when there are a plurality of objects with different attributes in the space in the vehicle traveling direction, and among the calculated plurality of candidates, the entrance is the most entrance in the parking direction The parking position determination device according to any one of claims 1 to 7, wherein the target stop position is determined based on the candidate on the side.
The parking position determination device according to any one of claims 1 to 8, wherein the camera is attached to a vehicle so as to image the rear of the vehicle.
Acquire an image of a camera that captures the surroundings of the vehicle,
From the storage unit that stores distance information representing the distance from the vehicle reference to the object in the vehicle longitudinal direction in association with the attribute for each attribute of the object, the object exists in the space in the vehicle traveling direction based on the image Acquiring the distance information associated with the attribute of the object;
The parking position determination program which makes a computer perform processing which determines the target stop position which concerns on the vehicle front-back direction at the time of parking completion using the acquired said distance information.