Parking assistance device

Abstract

Provided is a parking assistance device capable of appropriately identifying a parking zone of a vehicle. A parking assistance device (1) that identifies a parking zone in which a vehicle is parked includes: at least three cameras (10) that capture an outside of the vehicle; a setting unit (20) that sets a camera (10) that captures an image used to identify the parking zone, based on a setting condition that is set in advance; and a detection unit (30) that detects the parking zone based on an image captured by the set camera (10).

Description

TECHNICAL FIELD

[0001] The present application relates to a parking assistance device that recognizes a parking area in which a vehicle is parked. BACKGROUND

[0002] In the past, when parking a vehicle, a parking assistance device that assists a parking operation has been used in order to reduce the burden on the driver. When performing such an assisted parking operation, there are detection assistance that detects a parking area in which the vehicle can be parked, setting assistance that sets a target parking area in which the host vehicle is parked based on the detected parking area, start position guidance assistance that guides the vehicle to a parking start position at which parking is started, and parking position guidance assistance that guides the vehicle from the parking start position to the target parking area. In this series of assistance, it is important to recognize the parking area. As a technique related to such recognition of the parking area, for example, there is the technique described in Patent Literature 1 shown in the following citation.

[0003] In Patent Literature 1, a parking position information acquisition device that recognizes the position of a parking frame line with respect to the host vehicle based on an image of a camera mounted on the vehicle is described. The parking position information acquisition device detects the image of the parking frame line from a camera image captured by a camera provided on the side portion of the vehicle, and calculates the position of the parking frame line with respect to the position of the camera. Further, the position of the vehicle with respect to the position of the parking frame line is calculated as the own position.

[0004] Patent Literature 1: Japanese Patent Application Publication No. 2019-15570

[0005] The technique described in Patent Literature 1 configured as described above uses the position of the parking frame line for the calculation of the own position, in which the position of the parking frame line is obtained based on information of the camera image when the image of the parking frame line is close to the center in the region of the camera image. Here, a case where parking is performed to a parking area located in front of the vehicle is also assumed depending on the situation. The technique described in Patent Literature 1, since it uses the camera provided on the side portion of the vehicle as described above, although it is possible to appropriately recognize the parking area on both side portions of the vehicle, it does not assume a case where the parking area in front of the vehicle is recognized. Therefore, for example, it is not easy to recognize the parking area in front of the vehicle that advances into the garage, and it takes time to recognize.

[0006] Therefore, there is a need for a parking assistance device that can appropriately recognize the parking area of the vehicle. SUMMARY

[0007] The parking assistance device of the present application is characterized by the point that a parking area in which a vehicle is parked is recognized, the parking assistance device includes: at least three cameras that capture the outside of the vehicle; a setting unit that sets the cameras that capture an image used for recognizing the parking area based on a setting condition set in advance; and a detection unit that detects the parking area based on the image captured by the set cameras.

[0008] According to such a configuration, the image used for detecting the parking area can be set based on the setting condition, and the computational load involved in detecting the parking area or the detection accuracy can be improved as needed. Therefore, for example, by setting the setting condition according to the state of the vehicle, the parking area can be appropriately recognized according to the state of the vehicle.

[0009] Further, it is preferable that the setting condition be set according to the vehicle speed of the vehicle.

[0010] According to such a configuration, the image used for detecting the parking area can be set according to the vehicle speed of the vehicle.

[0011] Further, it is preferable that the detection unit detect a division line that defines the parking area.

[0012] According to such a configuration, since the parking area can be detected based on the division line, the detection accuracy of the parking area can be improved. Therefore, the parking area can be more appropriately recognized.

[0013] Further, it is preferable that the setting condition be set according to the parking method in which the vehicle is parked in the parking area and the unloading method in which the vehicle is unloaded from the parking area.

[0014] The part of the parking area that should be focused on differs depending on the situation when the vehicle is parked in the parking area or unloaded from the parking area. Therefore, according to the above configuration, since the image used for detecting the parking area is set according to the parking method or the unloading method of the vehicle, the part of the parking area that should be focused on according to the situation can be recognized.

[0015] Further, it is preferable that the setting unit be able to set a first mode in which both sides of the vehicle are set as a detection region of the parking area and a second mode in which another region different from the both sides is set as a detection region together with the both sides.

[0016] According to such a configuration, the first mode is set in a case where the parking area of the both sides of the vehicle is detected, and the second mode is set in a case where the parking area of another region together with the both sides of the vehicle is detected, so that detection corresponding to the state of the vehicle can be performed. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 This is a block diagram showing the configuration of a parking assist device.

[0018] Figure 2 is a schematic diagram of a vehicle equipped with a parking assist system.

[0019] Figure 3 This is an example of setting a condition.

[0020] Figure 4 (a) is an illustration of the left and right search mode. Figure 4 (b) is an illustration of the omnidirectional search mode.

[0021] Figure 5 This diagram shows the relationship between parking methods and exit methods, and the use of cameras that capture images.

[0022] Description of Reference Numerals

[0023] 1 ...parking assist device, 2 ...vehicle, 10 ...camera, 20 ...setting unit, 30 ...detection unit. DETAILED DESCRIPTION

[0024] The parking assistance device of the present invention is configured to appropriately identify the parking area where the vehicle is parked, regardless of the vehicle's driving conditions. Here, the so-called parking area in this embodiment is, for example, a space separated by dividing lines in a parking lot or other place so that vehicles can be parked one by one. The so-called dividing lines refer to signs or objects set on the road surface (ground). Specifically, they correspond to signs such as lines attached to the road surface, or obstacles, curbs, fences, etc. (hereinafter referred to as "obstacles, etc."). Therefore, the so-called parking area corresponds to a space separated by signs or objects set on the road surface in a parking lot or other place in order to park vehicles one by one. The present parking assistance device is configured to appropriately identify the parking area separated by such signs, obstacles, etc. The parking assistance device 1 of this embodiment is described below.

[0025] Figure 1 : is a block diagram showing the structure of the parking assist device 1 according to this embodiment. Figure 1 As shown, the parking assist device 1 includes various functional units such as a camera 10, a setting unit 20, and a detection unit 30. These functional units are constructed by hardware, software, or both, with a CPU as a core component, to perform processing related to parking space recognition.

[0026] Figure 2 Schematic diagram of a vehicle 2 equipped with the parking assist device 1. Here, for ease of understanding, in this embodiment, unless otherwise specified, the "front" ( Figure 2The direction of the arrow F shown in FIG. 1 means the front in the front-rear direction (driving direction) of the vehicle 2, and the rear ( Figure 2 The direction of the arrow B shown in FIG2 means the rear direction in the front-rear direction (driving direction) of the vehicle 2. In addition, the left-right direction or the lateral direction means the transverse direction of the vehicle 2 (vehicle width direction) perpendicular to the front-rear direction of the vehicle 2, that is, the left ( Figure 2 direction of the arrow L shown) and "right" ( Figure 2 The directions of the arrows R shown in the figure respectively indicate the left and the right of the vehicle 2 .

[0027] like Figure 2 As shown, at least three cameras 10 are provided in the vehicle 2. In the present embodiment, four cameras 10 are provided in the vehicle 2. Specifically, a front side camera 11 is provided in the front center portion of the vehicle 2 (e.g., the front grille), and a rear side camera 12 is provided in the rear center portion of the vehicle 2 (e.g., the trim panel). In addition, a left side camera 13 is provided at the left end portion of the vehicle 2 (e.g., the left rearview mirror 3), and a right side camera 14 is provided at the right end portion of the vehicle 2 (e.g., the right rearview mirror 4). Hereinafter, when the cameras 11-14 are not distinguished, they are collectively referred to as cameras 10, and when the cameras are distinguished, the respective reference numerals are used for description.

[0028] The camera 10 captures the exterior of the vehicle 2. The front camera 11 captures the scene within a shooting range (front shooting range 11A) set to include the road surface in front of the vehicle 2. The rear camera 12 captures the scene within a shooting range (rear shooting range 12A) set to include the road surface behind the vehicle 2. The left camera 13 captures the scene within a shooting range (left shooting range 13A) set to observe the road surface in front of the direction of travel of the vehicle 2 and to include the road surface on the left side. The right camera 14 captures the scene within a shooting range (right shooting range 14A) set to include the road surface on the right side of the vehicle 2. The cameras 11-14 can be configured so that the shooting ranges of the cameras adjacent to each other in the circumferential direction of the vehicle 2 overlap with each other, thereby reducing the area that becomes a blind spot.

[0029] In this embodiment, the camera 10 captures the exterior of the vehicle 2 via a fisheye lens. Therefore, the captured image in this embodiment is a so-called fisheye image. The captured image captured by the camera 10 is transmitted as image data to the detection unit 30 described later.

[0030] The setting section 20 sets the cameras 10 that capture the captured images used for recognition of the parking division based on the set conditions set in advance. In order to smoothly and appropriately recognize the parking division, the present parking assist device 1 is configured to be able to select the captured images that are the recognition processing targets (used for the recognition processing) in accordance with the situation of the vehicle 2. In the present embodiment, the conditions for selecting such captured images are set in advance as the set conditions in accordance with the situation of the vehicle 2, and are stored to the setting section 20. The "captured images used for recognition of the parking division" are captured images used for detection of the parking division by the detection section 30 described later.

[0031] For example, the set conditions can be set in accordance with the vehicle speed of the vehicle 2. Here, in the case of parking the vehicle 2, the case of parking to the parking division aligned (arranged in parallel with the traveling direction) along the traveling direction of the vehicle 2 is more frequent. In addition, in the case of searching for such a parking division, the case where the speed of the vehicle 2 is faster than in the process of parking to the parking division is more frequent. Therefore, in the case where the vehicle speed is equal to or higher than a predetermined speed, the setting section 20 sets the left camera 13 and the right camera 14 as the cameras 10 that capture the captured images used for recognition of the parking division. Thereby, in the left-right search mode, it is possible to ensure the detection accuracy of the parking division on the left and right sides of the vehicle 2, and to satisfy the demand of the occupant (see Figure 3 , for example, for the most frequent one-side parking in which the vehicle 2 is backed up while being parked to the parking division, it is possible to improve the convenience.

[0032] On the other hand, in the case where it is difficult to find the parking division for parking the vehicle 2, and it is finally necessary to find the parking division, it is more effective to search for the parking division not only along the traveling direction of the vehicle 2 (in parallel with the traveling direction), but also around the vehicle 2. In addition, in such a situation, the case where the speed of the vehicle 2 is reduced to search for the parking division is more frequent. Therefore, in the case where the vehicle speed is lower than the predetermined speed, the setting section 20 sets the front camera 11, the rear camera 12, the left camera 13, and the right camera 14 as the cameras 10 that capture the captured images used for recognition of the parking division. Thereby, by recognizing the parking division in the all-around search mode using all of the four cameras 10, it is possible to detect not only the division lines on the left and right sides of the vehicle 2, but also the division lines in front of and behind the vehicle 2, to recognize the parking division (see Figure 3 ), and it is possible to assist the parking operation, and also to cope with the front-in parking.

[0033] Accordingly, the setting unit 20 can set a left-right search mode (one example of the "first mode") and an all-around search mode (one example of the "second mode"), in which the left-right search mode sets both sides of the vehicle 2 as the detection area of the parking space; the all-around search mode sets, as the detection area, an area other than the both sides (in this embodiment, an area in the front-rear direction) together with the both sides. In Figure 4 In (a) of FIG. 8, a situation of the left-right search mode in which the left side photographing range 13A of the left side camera 13 and the right side photographing range 14A of the right side camera 14 of the vehicle 2 are set as the detection area of the parking space is shown. In Figure 4 In (b) of FIG. 8, a situation of the all-around search mode in which the front side photographing range 11A of the front side camera 11, the rear side photographing range 12A of the rear side camera 12, the left side photographing range 13A of the left side camera 13, and the right side photographing range 14A of the right side camera 14 of the vehicle 2 are set as the detection area of the parking space is shown.

[0034] In addition, the setting condition can also be set in accordance with a parking method of parking the vehicle 2 to the parking space and an unloading method of unloading the vehicle 2 from the parking space. The parking method of parking the vehicle 2 to the parking space is classified according to the parking of the vehicle 2 to the parking space, and corresponds to: (1) parallel parking by backing up the vehicle 2, (2) parallel parking by advancing the vehicle 2, (3) longitudinal parking by backing up the vehicle 2, and (4) longitudinal parking by advancing the vehicle 2. In addition, the unloading method of unloading the vehicle 2 from the parking space is classified according to the unloading of the vehicle 2 from the parking space, and corresponds to: (5) parallel parking by advancing the vehicle 2, (6) parallel parking by backing up the vehicle 2, (7) longitudinal parking by advancing the vehicle 2, and (8) longitudinal parking by backing up the vehicle 2.

[0035] In Figure 5 Examples of setting the cameras 10 that photograph the captured images for recognition of the parking space in accordance with the above-described parking method and unloading method are shown in Figure 5In the example of FIG. 6, (1) in the parallel parking backward entry, the captured image of each of the rear camera 12, the left camera 13, and the right camera 14 is set to be used, (2) in the parallel parking forward entry, the captured image of each of the front camera 11, the left camera 13, and the right camera 14 is set to be used, (3) in the vertical parking backward entry, the captured image of each of the front camera 11, the rear camera 12, the left camera 13, and the right camera 14 is set to be used, and (4) in the vertical parking forward entry, the captured image of each of the front camera 11, the rear camera 12, the left camera 13, and the right camera 14 is set to be used.

[0036] In addition, (5) in the parallel parking forward exit, the captured image of each of the front camera 11, the left camera 13, and the right camera 14 is set to be used, (6) in the parallel parking backward exit, the captured image of each of the rear camera 12, the left camera 13, and the right camera 14 is set to be used, (7) in the vertical parking forward exit, the captured image of each of the front camera 11, the rear camera 12, the left camera 13, and the right camera 14 is set to be used, and (8) in the vertical parking backward exit, the captured image of each of the front camera 11, the rear camera 12, the left camera 13, and the right camera 14 is set to be used. Of course, Figure 5 The set condition illustrated in FIG. 6 is one example, and a different set condition can be set. Figure 5

[0037] The setting unit 20 sets the cameras 10 that capture the captured images used for the recognition of the parking division based on such a set condition. The "setting the cameras 10 that capture the captured images used for the recognition of the parking division" means "setting the cameras 10 that capture the captured images used for the recognition of the parking division in the cameras 10", and in the present embodiment, means "setting the captured images used for the recognition of the parking division in each of the captured images captured by the cameras 10". The setting result of the setting unit 20 is at least transmitted to the detection unit 30 described later.

[0038] ​The detection unit 30 detects the parking zone on the basis of the set captured image captured by the camera 10. The set captured image captured by the camera 10 is the captured image captured by the camera 10 set by the setting unit 20. Here, the configuration can be such that the captured image (image data) is transmitted to the detection unit 30 only from the camera 10 set by the setting unit 20, or the configuration can be such that the captured image (image data) is transmitted to the detection unit 30 from all the cameras 10, and the detection unit 30 detects the parking zone using only the captured image (image data) transmitted from the camera 10 set by the setting unit 20. The camera 10 not set by the setting unit 20 can be configured to stop the capturing state or continue the capturing. In the case where the camera 10 not set by the setting unit 20 is configured to stop the capturing state, the setting result of the setting unit 20 is transmitted to the camera 10, and the capturing state is stopped on the basis of the setting result.

[0039] The detection unit 30 can be configured to, when detecting the parking zone, first detect a zone line that defines the parking zone, and detect the parking zone on the basis of the zone line. Such detection of the zone line can be performed using a publicly known method. Specifically, the detection unit 30 sets a detection area that serves as an area in which an end portion of the zone line is detected, along the road surface included in the captured image captured by the camera 10 set by the setting unit 20. The detection area is preferably set to be longer than the vehicle length in the vehicle length direction of the vehicle 2 and longer than the vehicle width in the vehicle width direction of the vehicle 2. The detection unit 30 detects the end portion of the sign, the obstacle, or the like in the zone line within the detection area set on the captured image. Although detailed description is omitted, for example, the detection can be performed by performing edge detection by a publicly known edge extraction in the detection area, and determining whether it is the sign, the obstacle, or the like on the basis of the detected edge.

[0040] In the present parking assistance device 1, as described above, the camera 10 that captures the captured image used for recognition of the parking zone is set, and the recognition of the parking zone is performed. Thus, since the captured image corresponding to the situation of the vehicle 2 can be used, the recognition accuracy can be improved, and the recognition speed can also be improved. Therefore, the parking zone can be recognized promptly and appropriately, and thus the parking assistance can be performed smoothly.

[0041] [Other Embodiments]

[0042] In the above embodiment, the case where four cameras 10 are provided is described as an example, but at least three cameras 10 can be provided in the vehicle 2. That is, three cameras 10 can be provided in the vehicle 2, or more than five cameras 10 can be provided. In either case, it is only necessary to provide the left camera 13 and the right camera 14.

[0043] In the above-described embodiment, the setting condition is set in accordance with the vehicle speed of the vehicle 2, but the setting condition can be set regardless of the vehicle speed of the vehicle 2.

[0044] In the above-described embodiment, the detection by the detection section 30 is described as detecting the division line of the prescribed parking division, but the detection section 30 can not detect the parking division based on the division line, and can be configured to directly detect the parking division.

[0045] In the above-described embodiment, the setting condition is set in accordance with the parking method of parking the vehicle 2 in the parking division and the unloading method of unloading from the parking division, but the setting condition can be set only in accordance with the parking method of parking the vehicle 2 in the parking division, or can be set only in accordance with the unloading method of unloading from the parking division. In addition, the setting condition can be set regardless of the parking method and the unloading method.

[0046] In the above-described embodiment, the setting section 20 can be configured to set the left-right search mode (first mode) in which both sides of the vehicle 2 are set as the detection area of the parking division, and the all-around search mode (second mode) in which an area other than the both sides is set as the detection area together with the both sides is described. For example, the first mode can be configured as the left-right search mode, and the second mode can be configured not to set the entire vehicle 2 as the detection area, but to set the left and right sides and the front of the vehicle 2 as the detection area, or to set the left and right sides and the rear as the detection area.

[0047] The present application can be used for a parking assistance device that identifies a parking division in which a vehicle is parked.

Claims

1. A parking assist device that identifies the parking area where the vehicle is parked. The parking assist device includes: at least three cameras to photograph the exterior of said vehicle; a setting unit that sets the camera that captures an image used to identify the parking area based on a preset setting condition; and The detection unit detects the parking area based on the image captured by the camera. When searching for the parking area, if the vehicle speed is greater than a predetermined speed, the setting unit sets the mode to the first mode, and if the vehicle speed is less than the predetermined speed, the setting unit sets the mode to the second mode. The first mode sets both sides of the vehicle as the detection area of ​​the parking area, and the second mode sets other areas different from the both sides as the detection area together with the both sides.

2. The parking assist device according to claim 1, wherein: The detection unit detects a partition line, wherein the partition line defines the parking zone.

3. The parking assist device according to claim 1 or 2, wherein: When the vehicle is parked in the parking area and when it is unparked from the parking area, the setting conditions are set according to a parking method for the vehicle in the parking area and a unparking method for the vehicle from the parking area.

Images