WO2017072955A1

WO2017072955A1 - Parking assistance device and parking assistance method

Info

Publication number: WO2017072955A1
Application number: PCT/JP2015/080746
Authority: WO
Inventors: 裕史狩田; 裕樹境; 下谷　光生; 井上　悟; 佐藤　理朗; 直志宮原
Original assignee: 三菱電機株式会社
Priority date: 2015-10-30
Filing date: 2015-10-30
Publication date: 2017-05-04
Also published as: JPWO2017072955A1; JP6509361B2; CN108140313A

Abstract

The purpose of the present invention is, with regard to parking assistance which guides a driver to an optimal parking space, to cause the driver to comprehend the reasons why a specific parking space is the optimal parking space. This parking assistance device 101 comprises: a photographic image acquisition unit 11 which acquires, from a camera which is mounted on a vehicle, a photographic image of the vicinity of the vehicle; an available parking space identification unit 13 which identifies available parking spaces in the photographic image; a guide path calculation unit which calculates guide paths for use in parking the vehicle in the available parking spaces; a parking difficulty assessment unit which assesses, as a degree of parking difficulty, the difficulty of parking in the available parking spaces by way of the guide paths; a recommended parking space identification unit which identifies a recommended parking space from among the available parking spaces; and a parking space guidance unit which supplies guidance to the driver of the vehicle to the recommended parking space, together with the degree of parking difficulty which is associated with the guide path.

Description

Parking assistance device and parking assistance method

This invention relates to a technique for supporting parking of a vehicle.

In recent years, parking assist devices that support parking operations have been used so that even beginners who are unfamiliar with parking can perform parking operations with peace of mind. The parking assistance device selects an optimal parking space from a plurality of parking spaces and teaches it to the driver, or teaches a guidance route to the optimal parking space.

For example, Patent Document 1 discloses a parking assist device that calculates the presence / absence of contact and the number of steerings for each parking space, and teaches the driver a parking space where there is no contact and the number of steerings is minimized. . Patent Document 2 discloses a method of calculating the presence / absence of contact and the number of steering operations in the parking assistance device of Patent Document 1.

JP-A-6-187596 Japanese Patent Laid-Open No. 5-2422

Since the processing of parking assistance devices is becoming more complex and sophisticated, it is important to properly teach the driver of the details of the processing for proper cooperation between the parking assistance system and the driver, and thus to realize secure parking. It is.

Current parking assistance is often used in semi-automatic driving in which a driver performs parking operation based on parking guidance to an optimal parking space taught by a parking assistance device. Therefore, it is expected that the driver will be instructed with a sufficient amount of information in an easy-to-understand manner why the parking space indicated by the parking assistance device is determined to be optimal.

However, in Patent Document 1, since it is only instructed by voice where the parking space should be parked, the driver cannot sufficiently tell why the parking space should be parked. There was a problem that it was difficult to judge whether or not parking could be carried out in accordance with the intentions.

The present invention has been made in view of the above-described problems, and it is an object of the present invention to make a driver understand why a specific parking space is optimal in parking assistance that teaches a driver the optimal parking space. .

A parking assist apparatus according to the present invention includes a captured image acquisition unit that acquires a captured image around a vehicle from a camera mounted on the vehicle, a parkable space specifying unit that specifies a parking space from the captured image, and parks the vehicle. A recommended route calculation unit that calculates a guidance route for parking in an available space, a parking difficulty level determination unit that determines the difficulty level of parking in a parking space by a guidance route as a parking difficulty level, and recommended parking from a parking space A recommended parking space specifying unit that specifies a space, and a parking space teaching unit that teaches the recommended parking space to the driver of the vehicle together with the degree of parking difficulty by the guidance route.

A parking assist method according to the present invention acquires a captured image around a vehicle from a camera mounted on the vehicle, specifies a parking space from the captured image, and calculates a guide route for parking the vehicle in the parking space. The parking difficulty level in the parking space by the guidance route is determined as the parking difficulty level, the recommended parking space is identified from the parking space, and the recommended parking space is set to the vehicle driver together with the parking difficulty level by the guidance route. Teach.

A parking assist apparatus according to the present invention includes a captured image acquisition unit that acquires a captured image around a vehicle from a camera mounted on the vehicle, a parkable space specifying unit that specifies a parking space from the captured image, and parks the vehicle. A recommended route calculation unit that calculates a guidance route for parking in an available space, a parking difficulty level determination unit that determines the difficulty level of parking in a parking space by a guidance route as a parking difficulty level, and recommended parking from a parking space A recommended parking space specifying unit that specifies a space, and a parking space teaching unit that teaches the recommended parking space to the driver of the vehicle together with the degree of parking difficulty by the guidance route. Therefore, the driver can determine whether or not to park in the recommended parking space with reference to the parking difficulty level.

A parking assist method according to the present invention acquires a captured image around a vehicle from a camera mounted on the vehicle, specifies a parking space from the captured image, and calculates a guide route for parking the vehicle in the parking space. The parking difficulty level in the parking space by the guidance route is determined as the parking difficulty level, the recommended parking space is identified from the parking space, and the recommended parking space is set to the vehicle driver together with the parking difficulty level by the guidance route. Teach. Therefore, the driver can determine whether or not to park in the recommended parking space with reference to the parking difficulty level.

The objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description and the accompanying drawings.

It is a block diagram which shows the structure of the parking assistance apparatus which concerns on Embodiment 1, and its periphery. It is a flowchart which shows the whole operation | movement of a parking assistance apparatus. It is a figure which shows the hardware constitutions of a parking assistance apparatus. It is a figure explaining the detection method of parking available space. It is a figure explaining the detection method of parking available space. It is a figure explaining the detection method of parking available space. It is a flowchart which shows the evaluation process of parking available space. It is a figure explaining the calculation method of the guidance route to the parking possible space. It is a figure explaining the calculation method of the frequency | count of steering. It is explanatory drawing of parking difficulty. It is a flowchart which shows the specific process of recommended parking space. It is a figure which shows a head unit and an instrument panel. It is a figure which shows the example of a display of a recommended parking space. It is a figure which shows the preparation method of the bird's-eye view displayed on an instrument panel. It is a figure which shows the example of a display when a driver | operator selects another parking space. The block diagram of the parking assistance apparatus comprised by the combination of the vehicle equipment and the server is shown. It is a block diagram which shows the parking assistance apparatus which concerns on Embodiment 2, and the structure of the periphery. It is a figure which shows the image of a preference database. It is a flowchart which shows the whole operation | movement of a parking assistance apparatus. It is a flowchart which shows the acquisition process of a driver's preference information by a personal authentication part. It is a figure which shows the registration screen of a driver | operator's preference information. It is a figure which shows the registration screen of a driver | operator's preference information. It is a flowchart which shows the calculation process of the preference matching degree by a preference matching degree calculation part. It is a flowchart which shows the outline | summary of the recommended parking space specific process by a recommended parking space specific part. It is a flowchart which shows the selection process of a recommended parking space 1st candidate. It is a flowchart which shows the selection process of a recommended parking space 2nd candidate. It is a flowchart which shows the selection process of a recommended parking space. It is a figure which shows the example of teaching of the recommended parking space in a parking space teaching part. It is a figure which shows the example of teaching of the recommended parking space in a parking space teaching part. It is a block diagram which shows the parking assistance apparatus which concerns on the modification of Embodiment 2, and the structure of the periphery. It is a block diagram which shows the parking assistance apparatus which concerns on the modification of Embodiment 2, and the structure of the periphery.

<A. Embodiment 1>
<A-1. Overall configuration>
FIG. 1 is a block diagram illustrating a configuration of the parking support apparatus 101 according to the first embodiment and its surroundings, and FIG.

The parking assistance device 101 includes a captured image acquisition unit 11, a distance calculation unit 12, a parking space identification unit 13, a parking space evaluation unit 14, a recommended parking space identification unit 17, and a parking space teaching unit 18.

The captured image acquisition unit 11 acquires the captured image of the image capturing unit 41 (step S11) and outputs it to the distance calculation unit 12. Here, the image photographing unit 41 is a camera using a CCD area sensor, for example, and is mounted on the vehicle 30 to photograph the rear of the vehicle 30.

The distance calculating unit 12 calculates the distance data from the captured image of the image capturing unit 41 (the captured image of the predetermined area including the parking space) to the object, and transmits the distance data to the parking space specifying unit 13.

The parking space specifying unit 13 specifies a parking space using the distance data calculated by the distance calculation unit 12 (step S12), and outputs information related to the parking space to the parking space evaluation unit 14.

The parking space evaluation unit 14 evaluates each parking space (step S13), and outputs the evaluation result to the recommended parking space identification unit 17. The parking space evaluation unit 14 includes a guidance route calculation unit 15 that calculates a guidance route to the parking space and a parking difficulty determination unit 16 that determines the parking difficulty of the parking space. The calculation of the guidance route and the determination of the parking difficulty level.

The recommended parking space specifying unit 17 specifies an optimum parking space as a recommended parking space based on the evaluation result of each parking space (step S14), and outputs information about the recommended parking space to the parking space teaching unit 18. .

The parking space teaching unit 18 teaches a recommended parking space (step S15).

FIG. 3 shows a hardware configuration of the parking support apparatus 101. As illustrated in FIG. 3, the parking assistance device 101 is realized by a processor 51, a memory 52, and a reception unit 53. The distance calculation unit 12, the parking space identification unit 13, the parking space evaluation unit 14, the recommended parking space identification unit 17, and the parking space teaching unit 18 are stored in a memory 52 such as a RAM (Random Access Memory). By executing the program, the function of the processor 51 is realized. However, these may be realized in cooperation with a plurality of processors 51, for example. The captured image acquisition unit 11 is realized by the reception unit 53.

<A-2. Configuration of each part>
The distance calculation unit 12 calculates distance data to the object for each direction from the image data of the captured image acquired by the captured image acquisition unit 11, and transmits the distance data to the parking space specifying unit 13. The calculated distance data is expressed by data of distance R and azimuth θ. In addition, ECU is assumed as a device for distance calculation.

The parking space specifying unit 13 converts the distance data received from the distance calculation unit 12 into an xy orthogonal coordinate system. The x-axis and the y-axis are orthogonal to each other, and the origin of these axes is the center of the field angle of the CCD area sensor.

4 to 6 are diagrams for explaining the parking space specifying process (step S12 in FIG. 2) by the parking space specifying unit 13. As shown in FIG. 4, when the vehicle is already parked in some of the parking spaces of the parking lot, the parking space specifying unit 13 specifies a vacant space in the parking space, that is, a parking available space. The distance data in the xy orthogonal coordinate system indicates the distance between the vehicle and the already parked vehicle. The parking space specifying unit 13 performs the least square method on the distance data to obtain x = A ₁ · y + B ₁ . Further, the second least square method is performed on the distance data satisfying x <A ₁ · y + B ₁ to obtain x = A ₂ · y + B ₂ , which is defined as the parking space front boundary line.

Next, left and right edges are detected by the method shown in FIG. 5 with respect to the front boundary line of the parking space. Specifically, the distance Δx _i between the distance data to be drawn and the parking space front boundary line is expressed by Expression (1).

Here, x _i = R _i · cos (θ _i ) and y _i = R _i · sin (θ _i ). Next, when the threshold Th is set to 0.6 m, for example, when the distance data is scanned and Δx _i-1 > Th and Δx _i ≤Th, the i-th coordinate is the right edge coordinate, Δx _i-1 ≤Th and If ΔX _i > Th, the i-th coordinate is detected as the left edge coordinate.

Next, the left and right edges of the parking space are projected onto the front boundary line of the parking space, and the center coordinates and width of the empty parking space are calculated (see FIG. 6). If the right edge coordinates after projection are (Gx ₁ (i), GY ₁ (i)) and the left edge coordinates after projection are (Gx ₂ (i), GY ₂ (i)), the center of each empty parking space The coordinates GX (i), GY (i), and the width W (i) can be calculated by equations (2) to (4).

Then, the width W (i) of the empty parking space is compared with the vehicle width WC. If W (i)> WC + β, the empty parking space is detected as a parking available space. If W (i) ≦ WC + β, it is determined that the empty parking space is not a parking available space. Note that β is a clearance.

The parking space evaluation unit 14 includes a guidance route calculation unit 15 that calculates a guidance route to each parking space and a parking difficulty determination unit 16 that determines the parking difficulty of each parking space. Then, each parking available space specified by the parking available space specifying unit 13 is evaluated.

FIG. 7 is a flowchart showing the evaluation process (step S13 in FIG. 2). First, the parking space evaluation unit 14 determines whether or not the evaluation has been completed for all parking spaces (step S131). If the parking space evaluation unit 14 has not completed the calculation, a guidance route is calculated for an unrated parking space (step S132). ), The number of steerings until the parking is completed is calculated (step S133).

The guidance route calculation unit 15 calculates the guidance route to each parking space using the state equation of Equation (5).

Note that α represents the steering angle of the vehicle, L represents the wheel base of the vehicle 30, (x, y) represents the position of the vehicle, and θ represents the attitude angle of the vehicle (see FIG. 8). In this equation, the number of times α is updated from the initial state to the guidance completion state corresponds to the number of steering operations (see FIG. 9).

Next, the parking difficulty level determination unit 16 calculates the parking difficulty level of each parking space (step S134). FIG. 10 is an explanatory diagram of the parking difficulty level. FIGS. 10 (a) and 10 (b) show a guide route for parking in a parking space. In any of the examples shown in FIGS. 10 (a) and 10 (b), three times Parking is completed with the number of steerings. However, in the case of FIG. 10A, since the moving distance between each steering is shorter than in the case of FIG. 10B, a precise parking operation is required. More specifically, an error from the simulation increases due to excessive depression of the accelerator or a delay in the depression of the brake, and as a result, there is a high possibility of parking failure, and therefore, it can be said that the parking difficulty is high.

Therefore, the parking difficulty level determination unit 16 calculates the parking difficulty level evaluation value C based on the sum of the reciprocals of the moving distances between the steerings (formula (6)).

N represents the number of times of steering, and dist (i) represents the moving distance from the i-th steering to the i + 1-th steering. The parking difficulty level evaluation value C indicates a larger value as the moving distance between each steering is shorter. When the number of times of steering is 3, C = 1 / (1 → second travel distance) + 1 / (2 → third travel distance).

Here, since the vehicle stop position at the first steering differs depending on the driver, it is considered that the moving distance from the steering 1 to the second includes a large error. Therefore, the steering 1 → second movement distance may be excluded from the calculation of the parking difficulty level C. In this case, the parking difficulty level evaluation value C is expressed by Expression (7).

When the evaluation (calculation of the guidance route, the calculation of the parking difficulty evaluation value) is completed for all the parking spaces, the parking space evaluation process ends.

FIG. 11 is a flowchart showing the recommended parking space specifying process (step S14 in FIG. 2) by the recommended parking space specifying unit 17. Hereinafter, the recommended parking space specifying process will be described with reference to FIG. First, the recommended parking space specifying unit 17 determines whether or not processing has been completed for all parking spaces (step S141). If there is an unprocessed parking space, the following processing is performed for the parking space. First, it is determined whether the parking difficulty level evaluation value C is less than a predetermined upper limit value (step S142). If the parking difficulty level evaluation value C is equal to or greater than the predetermined upper limit value, the process returns to step S141. If the parking difficulty level evaluation value C is less than the predetermined upper limit value, the upper limit value is updated with the parking difficulty level evaluation value C (step S143). And the said parking possible space is determined to be a recommended parking space (step S144). By performing this process for all the parking spaces, the parking space having the smallest parking difficulty evaluation value C is finally determined as the recommended parking space.

The parking space teaching unit 18 teaches a recommended parking space using the head unit 32 and the instrument panel 33 in front of the handle 31 shown in FIG. 12 (step S15 in FIG. 2). As shown in FIG. 13A, the head unit 32 displays a superposed display of the recommended parking space on the camera image behind the vehicle photographed by the image photographing unit 41. In addition, the parking difficulty level when parking in the recommended parking space along the guidance route is displayed together with the recommended parking space. Here, the parking difficulty level is divided into five levels from the parking difficulty level evaluation value C, and the parking difficulty level is shown in five levels. Therefore, the driver can determine whether or not to park in the recommended parking space with reference to the parking difficulty level. On the instrument panel 33, a bird's-eye view of the vehicle and the parking space is displayed as shown in FIG. This bird's-eye view shows parking spaces (1) and (2), of which parking space (1) is determined as a recommended parking space, and a guidance route there is shown. The driver can grasp the driving operation for parking in the recommended parking section by looking at this display.

In addition, the parking space teaching unit 18 detects from the speaker mounted on the vehicle “two spaces where parking is possible. (1) requires three steerings and the parking difficulty is 2. Unless there is a specific reason ( It is also possible to output a voice such as “I recommend parking at 1)”.

Note that the bird's-eye view image shown in FIG. 13B can be created based on the distance data obtained from the distance calculation unit 12, as shown in FIG. FIG. 14A shows distance data obtained from the distance calculation unit 12, and FIG. 14B shows a bird's-eye view used for display on the instrument panel 33.

Since the teaching is performed as described above, the driver can determine whether or not the recommended parking space should be parked after knowing the number of steerings of the recommended parking space and the parking difficulty. In the example of FIG. 13, when the driver wants to park in a place other than the recommended parking space, for example, the user can select the parking space (2) by an operation such as touching the screen or pressing a specific switch. Also good. FIG. 15 shows a screen display example in that case. FIG. 15A shows a display example of the head unit 32, and FIG. 15B shows a display example of the instrument panel 33. In this case, as shown in FIG. 15A, the parking space (2) is displayed as the selected parking space, and the number of steerings and the difficulty level of the parking space (2) are displayed on the head unit 32. Moreover, as shown in FIG.15 (b), the guidance path | route to the parking space (2) is displayed on the instrument panel 33 by a bird's-eye view. And from the speakers mounted on the vehicle, “We detected two spaces that can be parked. (2) requires three times of steering, and the distance between steering is short and requires careful operation. 3. If there is no particular reason, it is recommended to park to (1). "

<A-3. Modification>
In the above description, the distance calculation unit 12, the parking space identification unit 13, the parking space evaluation unit 14, the recommended parking space identification unit 17, and the parking space teaching unit 18 are stored in the memory 52 or the like in the processor 51 of FIG. 3. It was realized by operating according to the software program. However, instead of this, the distance calculation unit 12, the parking space identification unit 13, the parking space evaluation unit 14, the recommended parking space identification unit 17, and the parking space teaching unit 18 realize the operation by an electrical circuit of hardware. It may be realized by a signal processing circuit. Software distance calculation unit 12, parking space identification unit 13, parking space evaluation unit 14, recommended parking space identification unit 17 and parking space teaching unit 18, hardware distance calculation unit 12, parking space identification unit 13, As a concept combining the parking space evaluation unit 14, the recommended parking space specifying unit 17, and the parking space teaching unit 18, the word “processing circuit” can be used instead of the word “part”.

In FIG. 1, all the configurations of the parking support device 101 are described as being mounted on the vehicle 30, but each configuration is shared with a mobile terminal such as a smartphone or a PDA, a server, or the like in addition to the in-vehicle device. Also good. In FIG. 16, the block diagram of the parking assistance apparatus 101 comprised by the combination of the vehicle-mounted apparatus and the server is shown. In FIG. 16, the captured image acquisition unit 11 and the parking space teaching unit 18 are configured by an in-vehicle device, and the distance calculation unit 12, the parking space identification unit 13, the parking space evaluation unit 14, and the recommended parking space identification unit 17 are servers. It is configured.

<A-4. Effect>
The parking assist apparatus 101 according to the first embodiment includes a captured image acquisition unit 11 that acquires a captured image around the vehicle from a camera mounted on the vehicle, and a parkingable space specifying unit that specifies a parking space from the captured image. 13, a guidance route calculation unit 15 that calculates a guidance route for parking the vehicle in the parking space, and a parking difficulty determination unit 16 that determines the difficulty level of parking in the parking space by the guidance route as a parking difficulty level. And a recommended parking space specifying unit 17 for specifying a recommended parking space from the parking available space, and a parking space teaching unit 18 for teaching the recommended parking space to the driver of the vehicle together with the parking difficulty level by the guidance route. Therefore, the driver can determine whether or not to park in the recommended parking space based on the parking difficulty level.

Also, the parking difficulty level determination unit 16 calculates the sum of the reciprocal of the distance traveled between each steering until the vehicle is parked along the guidance route, and determines the parking difficulty level based on the total. The shorter the movement distance between each steering, the higher the possibility of parking failure due to a slight shift in the timing of the parking operation. In such a case, the parking difficulty can be determined higher.

Further, the parking difficulty determination unit 16 excludes the reciprocal of the moving distance between the first steering and the second steering along the guidance route from the calculation of the sum, thereby determining the vehicle stop position at the first steering. An error is excluded and the parking difficulty level can be determined with high accuracy.

In addition, when the image capturing unit 41 is a camera that captures the rear of the vehicle, since the captured image of the camera can be displayed on the head unit 32 when parking backward, it is easy to park while viewing the image.

In addition, the parking support method according to the first embodiment acquires a captured image around the vehicle from a camera mounted on the vehicle, specifies a parking available space from the captured image, and parks the vehicle in the parking available space. Calculate the guidance route, determine the parking difficulty level in the parking space by the guidance route as the parking difficulty level, identify the recommended parking space from the parking space, and identify the recommended parking space along with the parking difficulty level by the guidance route. Teach the driver. Therefore, the driver can determine whether or not to park in the recommended parking space based on the parking difficulty level.

<B. Second Embodiment>
<B-1. Configuration>
FIG. 17 is a block diagram illustrating a configuration of the parking assist device 102 according to the second embodiment. The parking assistance device 102 includes a personal authentication unit 19, a preference database 20, and a preference matching degree calculation unit 21 in addition to the configuration of the parking assistance device 101 according to the first embodiment.

An image of the preference database 20 is shown in FIG. Driver preference data is registered in the preference database 20. The driver preference data includes a set of a driver image (a face image or a fingerprint image), a driver name, and a driver preference (a plurality can be set and the higher the priority is set). As with a general database, data can be added, deleted, and modified, but the driver preference is limited to that which can be evaluated by the preference matching degree calculation unit 21 described later. The driver can freely select from the limited driver preference data.

Although a HDD (Hard Disk Drive) is assumed as a device for the preference database 20, a cloud or a server may be used instead of the device. The driver image data is assumed to be a bitmap, but a lossless compression image format such as PNG may be used for capacity reduction. Driver preference data can be expressed in integer form. As described above, since the driver preference data is limited, for example, the driver preference data is preliminarily set such as “I do not want to park next to a luxury car: 1” or “I want to park in the shade: 2”. This is because can be associated with an integer.

The personal authentication unit 19 identifies the driver and refers to the preference database 20 to acquire the driver's preference information.

The preference coincidence calculation unit 21 calculates the preference coincidence of each parking space using an image recognition algorithm based on the image data received from the captured image acquisition unit 11 and the preference information received from the personal authentication unit 19.

<B-2. Operation>
FIG. 19 is a flowchart showing the overall operation of the parking assistance apparatus 102. The parking assistance apparatus 102 acquires a captured image (step S21), specifies a parking space (step S22), and evaluates the parking space (step S23). These steps are the same as those of the parking assist apparatus 101 according to the first embodiment.

Next, the personal authentication unit 19 acquires driver preference information (step S24). FIG. 20 is a flowchart showing a process for acquiring driver preference information by the personal authentication unit 19. In acquiring the driver's preference information, the personal authentication unit 19 first acquires the driver information (step S241). As a method for acquiring driver information, authentication by image recognition such as face authentication or fingerprint authentication is common, but authentication by a SIM card built in a smartphone may be used. In the case of face authentication, a driver's face image may be captured by a camera 34 attached to the upper part of the instrument panel 33 and the captured image may be transmitted to the personal authentication unit 19. In the case of fingerprint authentication, a driver's fingerprint image is acquired by a fingerprint acquisition sensor attached to the handle 31, and the fingerprint image is transmitted to the personal authentication unit 19. In the case of authentication using the SIM card, the SIM card may be inserted into the SIM card insertion slot attached to the handle 31 and personal information may be transmitted to the personal authentication unit 19.

Next, the personal authentication unit 19 collates the driver information with the information registered in the preference database 20, and whether there is preference information that matches the driver information, that is, whether the driver preference information is registered. Is determined (step S242). If the driver's preference information is registered as a result of the collation, the preference information is acquired (step S244) and transmitted to the preference matching degree calculation unit 21. If the driver's preference information is not registered, the preference information is newly registered in the preference database 20 (step S243), and the preference information is acquired (step S244).

FIG. 21 shows a driver preference information registration screen displayed on the instrument panel 33. When the driver's preference information is already registered in the preference database 20, the preference information is displayed on the instrument panel 33. On the other hand, if the driver's preference information is not yet registered in the preference database 20, a screen prompting the registration of preference information is displayed on the instrument panel 33. For the driver's face photograph, a photographed image of the camera 34 installed above the instrument panel 33 is used.

Note that preference information cannot be registered in duplicate. If registration is attempted in duplicate, a pop-up display appears as shown in FIG. 22, and a warning is given to the driver.

Referring back to FIG. 19, in step S25, the preference matching degree calculation unit 21 calculates the preference matching degree for each parking space based on the preference information acquired in step S24. FIG. 23 is a flowchart showing the preference matching degree calculation processing by the preference matching degree calculation unit 21. The preference coincidence calculation unit 21 determines whether the preference coincidence has been calculated for all preference information (step S251).

If there is preference information for which the preference matching degree is not calculated, the preference matching degree is calculated for the preference information (step S252). The preference coincidence calculation unit 21 executes an image recognition algorithm on the captured image acquired by the captured image acquisition unit 11, and determines how much each parking available space matches the driver preference. Here, the reason why the image recognition is used for calculating the preference coincidence is that the human judges the preference mainly based on information from the visual sense. For example, if the driver preference is “I do not want to park next to a luxury car”, the preference matching degree is calculated by specifying the vehicle types of all detected vehicles. If the driver's preference is “I want to park in the shade”, a darker area than the surroundings is specified and the preference matching degree is calculated. If the driver's preference is “I do not want to park next to a large car”, the vehicle type plate is identified by recognizing the letters on the car's license plate, and the degree of preference matching is calculated. As described above, the preference matching degree calculation unit 21 is equipped with an image recognition algorithm for calculating the preference matching degree corresponding to each preference information.

The preference matching degree is normalized with a value of 0.0 to 1.0, and if 0.0, it is not matched at all, and if it is 1.0, it is assumed that it is completely matched. If the image recognition fails and the preference coincidence cannot be calculated (step S253: No), the preference coincidence is set to 99 (step S254).

The calculated preference coincidence is transmitted to the recommended parking space specifying unit 17, and when the preference coincidence calculation is completed for all preference information (step S251: Yes), the preference coincidence calculation process ends.

19, after calculating the preference coincidence, the recommended parking space specifying unit 17 specifies the recommended parking space (step S26). In the first embodiment, the recommended parking space specifying unit 17 specifies the one having the lowest parking difficulty as the recommended parking space. However, in the second embodiment, in addition to the parking difficulty, the preference matching degree and the number of steerings are considered. Identify recommended parking spaces.

FIG. 24 is a flowchart showing an outline of recommended parking space specifying processing by the recommended parking space specifying unit 17. Hereinafter, the recommended parking space specifying process will be described. First, the recommended parking space specifying unit 17 selects a parking space having a high preference coincidence as a recommended parking space first candidate (step S261). Next, the recommended parking space first candidate having the smallest number of steerings is selected as the recommended parking space second candidate (step S262). And the thing with the lowest parking difficulty from a recommended parking space 2nd candidate is determined to be a recommended parking space (step S263).

FIG. 25 is a flowchart showing the recommended parking space first candidate selection process (step S261 in FIG. 24). The recommended parking space first candidate selection process will be described below. The recommended parking space specifying unit 17 determines whether or not the selection process of the first recommended parking space candidate is completed for all parking spaces (step S2611). Perform the process. That is, it is determined whether or not the driver's preference information exists (step S2612). If the preference information does not exist, the available parking space is registered as a recommended parking space first candidate (step S2613), and the process returns to step S2611.

If the preference information exists, it is determined whether or not the preference matching degree determination process has been completed for all the existing preference information (step S2614). If the preference matching degree determination process is incomplete, the process advances to step S2615 and subsequent steps for the undetermined preference information.

In step S2615, it is determined whether the preference coincidence is equal to or greater than a threshold value. If the preference coincidence is equal to or greater than the threshold value, the available parking space is registered as a recommended parking space first candidate (step S2616). ). If the preference coincidence is less than the threshold, the process returns to step S2614. If there is other unprocessed preference information, the preference coincidence is determined for the preference information (step S2615). Therefore, when the preference coincidence of the parking available space is equal to or greater than the threshold for any preference information, the parking available space is selected as the recommended parking space first candidate.

When the above processing is completed for all the parking spaces (step S2611: Yes), the recommended parking space first candidate selection processing ends. With the above processing, when there is no preference information, all the parking spaces are selected as the recommended first recommended parking space, and when preference information exists, the preference matching degree for any preference information is greater than or equal to the threshold value. Is selected as the first recommended parking space candidate.

FIG. 26 is a flowchart showing the recommended parking space second candidate selection process (step S262 in FIG. 24). Hereinafter, the selection process of the recommended parking space second candidate selection process will be described. The recommended parking space specifying unit 17 determines whether or not the selection process of the second recommended parking space candidate has been completed for all the recommended first parking space candidates (step S2621). The following processing is performed for the first space candidate. That is, it is determined whether the number of times of steering of the recommended parking space first candidate is equal to or less than the upper limit value (step S2622). Here, the initial value of the upper limit value of the number of steerings is set in advance. If the number of steerings is less than or equal to the upper limit, the upper limit is updated with the number of steerings of the recommended parking space first candidate (step S2623), and the recommended parking space first candidate is selected as the recommended parking space second candidate (step S2624), the process returns to step S2621.

On the other hand, if the number of times of steering is equal to or greater than the upper limit, the process returns to step S2621. When the recommended parking space second candidate selection process is completed for all the recommended parking space first candidates, the recommended parking space second candidate selection process ends. With the above processing, the recommended parking space first candidate with the smallest number of steerings is selected as the recommended parking space second candidate.

FIG. 27 is a flowchart showing a recommended parking space selection process (step S263 in FIG. 24). Hereinafter, the recommended parking space specifying process will be described. The recommended parking space specifying unit 17 determines whether there is a second recommended parking space candidate (step S2631). If the second recommended parking space candidate does not exist, the process ends without selecting the recommended parking space. If there is a second recommended parking section candidate, it is determined whether or not the recommended parking space specifying process has been completed for all the recommended parking space second candidates (step S2632). The following processing is performed for the second parking space candidate. That is, it is determined whether the parking difficulty level evaluation value of the recommended parking space second candidate is less than the upper limit value (step S2633). Here, the initial value of the upper limit value of the parking difficulty level evaluation value is set in advance. If the parking difficulty level evaluation value is greater than or equal to the upper limit value, the process returns to step S2632. On the other hand, if the parking difficulty level evaluation value is less than the upper limit value, the upper limit value is updated with the parking difficulty level evaluation value of the recommended parking space second candidate (step S2634), and the recommended parking space second candidate is set as the recommended parking space. (Step S2635) and the process returns to step S2632.

In addition, the parking difficulty level evaluation value of each parking space is obtained by the parking difficulty level determination unit 16, and the details thereof are as already described in the first embodiment.

When the recommended parking space specifying process is completed for all the recommended parking space second candidates, the recommended parking space specifying process ends. With the above processing, the second recommended parking space candidate with the smallest parking difficulty evaluation value is specified as the recommended parking space.

28 and 29 are diagrams showing an example of teaching a recommended parking space in the parking space teaching unit 18. FIG. 28 shows teaching of a recommended parking space when the driver's preference is “I do not want to park next to a luxury car”. The parking space teaching unit 18 teaches a recommended parking space using the head unit 32 and the instrument panel 33 in front of the handle 31 (see FIG. 12). FIG. 28A shows a display screen of the head unit 32, and FIG. 28B shows a display screen of the instrument panel 33. On the instrument panel 33, a bird's eye view showing a recommended parking section, a guide route, and a vehicle is displayed. In FIG. 28 (b), the parking spaces are indicated by (1) and (2), but (2) does not match the driver's preference because a luxury car is parked next to it. Therefore, (1) is taught as a recommended parking space. The head unit 32 displays a camera image of the rear of the vehicle captured by the image capturing unit 41 and a recommended parking space superimposed thereon, and further displays the number of times of steering “3 times” and the parking difficulty level “3”. The In addition, the parking space teaching unit 18 has detected one parking space that matches the driver's preference of “I do not want to park next to a luxury car” from a speaker mounted on the vehicle. (1) requires three steerings. The parking difficulty is 3 because the moving distance before and after steering is short. As a result, the driver knows that the recommended parking space presented matches his own preference of “I do not want to park next to a luxury car” but the parking difficulty is high, and then the recommended parking space It can be determined whether or not to park.

FIG. 29 shows teaching of a recommended parking space when the driver's preference is “I want to park in the shade”. 28A shows the display screen of the head unit 32, and FIG. 29B shows the display screen of the instrument panel 33, as in FIG. In FIG. 29 (b), parking spaces are indicated by (1) and (2), but (1) is not shaded and therefore does not match the driver's preference. Therefore, (2) is taught as a recommended parking space. In addition, the head unit 32 displays a superposed display of the recommended parking space on the camera image captured by the image capturing unit 41, and further displays the number of steerings “3 times” and the parking difficulty level “2”. The In addition, the parking space teaching unit 18 has detected one parking space that matches the driver's preference of “I want to park in the shade” from a speaker mounted on the vehicle. (2) requires three steerings. The parking difficulty is “2”. As a result, the driver knows that the recommended parking space presented matches his preference of “I want to park in the shade” and the parking difficulty level is not high. Can be determined.

<B-3. Modification>
In the description so far, it has been described that the image capturing unit 41 is a camera that captures the rear of the vehicle 30. However, since the image capturing unit 41 only needs to capture the parking space, it may be a camera that captures the front or side of the vehicle 30 instead of the rear. In addition, when the image capturing unit 41 is a camera that captures the rear of the vehicle 30, the image captured by the camera can be displayed on the head unit 32 during parking in the rear, so that it is easy to park while viewing the image. There are advantages.

Further, as in the parking assistance device 103 shown in FIG. 30, the image capturing unit 41 may be a combination of a camera that captures the rear of the vehicle 30 and a camera that captures the front of the vehicle 30. Note that the arrows from the image capturing unit 41 to the captured image acquiring unit 41 added in FIG. 30 are not shown. In this case, the recommended parking space can be displayed on the head unit 32 by superimposing the information on the recommended parking space on the image ahead of the vehicle at the time of front parking. Space information can be presented to the driver. Moreover, there is an advantage that an image of a parking space can be acquired quickly and accurately by using a camera that shoots in multiple directions.

Further, like the parking assist device 104 shown in FIG. 31, the image capturing unit 41 may be a combination of a camera that captures the rear of the vehicle 30 and a camera that captures the side (right side, left side) of the vehicle 30. By photographing the side of the vehicle 30, a bird's-eye view of the instrument panel 33 can be created with high accuracy, and the preference matching degree by image recognition can be calculated with high accuracy.

<B-4. Effect>
In addition to the configuration of the parking support device 101 according to the first embodiment, the parking support device 102 according to the second embodiment includes a personal authentication unit 19 (preference acquisition unit) that acquires a preference regarding the parking space of the driver of the vehicle. , A preference matching degree calculation unit 21 that calculates a preference matching degree indicating how much the parking available space matches the preference based on the photographed image, and the recommended parking space specifying unit 17 can park based on the preference matching degree. Identify the recommended parking space from the space. Therefore, it is possible to teach a recommended parking space according to the driver's preference.

In addition, the recommended parking space specifying unit 17 specifies a recommended parking space from the available parking space where the preference matching degree is equal to or greater than the threshold value, and the preference matching is calculated when the preference matching degree calculation unit 21 fails to calculate the preference matching degree. Force the degree to a value above the threshold. Therefore, it is possible to specify the recommended parking space even when the preference coincidence cannot be calculated for reasons such as image recognition failure.

In addition, the parking space teaching unit 18 teaches the preference that has become the basis for the recommended parking space identifying unit 17 to identify the recommended parking space when teaching the recommended parking space to the driver of the vehicle. Therefore, the driver can determine whether or not to park in the recommended parking space after knowing the preference that is the basis for specifying the recommended parking space.

Also, the above preference includes at least one of being shaded and not having a specific vehicle model parked adjacent to each other, so that it is possible to teach a recommended parking space that matches such preference.

Further, the parking support device 102 includes a preference database 20 that stores in advance preferences related to the parking space of the driver of the vehicle. Then, the personal authentication unit 19 (preference acquisition unit) acquires the preference regarding the parking space of the driver of the vehicle from the preference database 20. Therefore, preference can be acquired without having the user input each time.

Further, when the image capturing unit 41 is a camera that captures the rear and front of the vehicle, the information on the recommended parking space can be superimposed on the image in front of the vehicle and displayed on the head unit 32 when the vehicle is parked forward. In either case of front parking or rear parking, the information on the recommended parking space can be presented to the driver in an easy-to-understand manner. Moreover, the image of the parking space can be acquired quickly and accurately by photographing multiple directions.

In addition, when the image capturing unit 41 is a camera that captures the rear of the vehicle, the bird's-eye view of the instrument panel 33 can be generated with high accuracy, and the preference matching degree by image recognition can be calculated with high accuracy. .

In the present invention, it is possible to freely combine the respective embodiments within the scope of the invention, and to appropriately modify and omit the respective embodiments.

Although the present invention has been described in detail, the above description is illustrative in all aspects, and the present invention is not limited thereto. It is understood that countless variations that are not illustrated can be envisaged without departing from the scope of the present invention.

11 Captured image acquisition unit, 12 Distance calculation unit, 13 Parking space identification unit, 14 Parking space evaluation unit, 15 Guide route calculation unit, 16 Parking difficulty determination unit, 17 Recommended parking space identification unit, 18 Parking space teaching unit , 19 personal authentication unit, 20 preference database, 21 preference coincidence calculation unit, 30 vehicle, 31 handle, 32 head unit, 33 instrument panel, 34 camera, 41 image capturing unit, 51 processor, 52 memory, 53 receiving

unit

101, 102, 103, 104 Parking assistance device.

Claims

A captured image acquisition unit that acquires a captured image around the vehicle from a camera mounted on the vehicle;
A parking space identifying unit that identifies a parking space from the captured image;
A guidance route calculation unit for calculating a guidance route for parking the vehicle in the parking space;
A parking difficulty determination unit that determines the difficulty of parking in the parking space by the guidance route as a parking difficulty;
Based on the number of steerings along the guidance route and at least one of the parking difficulty levels, a recommended parking space identifying unit that identifies a recommended parking space from the parking space,
A parking space teaching unit for teaching the recommended parking space to a driver of the vehicle together with a parking difficulty level by the guidance route;
Comprising
Parking assistance device.
The parking difficulty level determination unit calculates the sum of the reciprocal of the moving distance between each steering until the vehicle is parked along the guidance route, and determines the parking difficulty level based on the total.
The parking assistance device according to claim 1.
The parking difficulty level determination unit calculates the sum by excluding the reciprocal of the moving distance between the first steering and the second steering along the guidance route;
The parking assistance device according to claim 2.
A preference acquisition unit for acquiring preferences relating to the parking space of the driver of the vehicle;
A preference coincidence calculating unit that calculates a preference coincidence indicating how much the parkable space matches the preference based on the captured image;
The recommended parking space specifying unit specifies the recommended parking space from the parking available space based on the preference coincidence.
The parking assistance device according to claim 1.
The recommended parking space specifying unit specifies the recommended parking space from the parking available space where the preference coincidence is equal to or higher than a threshold, and when the preference coincidence calculation unit fails to calculate the preference coincidence, Forcibly set the preference match to a value greater than or equal to the threshold,
The parking assistance device according to claim 4.
In teaching the recommended parking space to the driver of the vehicle, the parking space teaching unit also teaches the preference that is the basis for the recommended parking space specifying unit to specify the recommended parking space.
The parking assistance device according to claim 4.
The preference includes at least one of being shaded and that a vehicle of a specific vehicle type is not adjacently parked.
The parking assistance device according to claim 6.
A preference database pre-stored with regard to the parking space of the driver of the vehicle;
The preference acquisition unit acquires the preference regarding the parking space of the driver of the vehicle from the preference database.
The parking assistance device according to claim 4.
The camera captures the rear of the vehicle;
The parking assistance device according to claim 1.
The camera also captures the front of the vehicle,
The parking assistance device according to claim 9.
The camera also captures the side of the vehicle,
The parking assistance device according to claim 9.
Obtain a captured image around the vehicle from a camera mounted on the vehicle,
Identify a parking space from the captured image,
Calculating a guidance route for parking the vehicle in the parking space;
Determining the difficulty of parking in the parking space by the guidance route as a parking difficulty,
Identify the recommended parking space from the parking space,
Teach the recommended parking space to the driver of the vehicle together with the parking difficulty level by the guidance route,
Parking assistance method.