Parking support device

Abstract

To provide a parking support device that can improve convenience of leaning-type automatic parking.SOLUTION: The parking support device comprises map generating means and running control means. The map generating means registers map including a route extending from a parking start position to a parking position, during running for learning. The running control means makes a vehicle autonomously run on the basis of the map, during automatic parking. The map includes a plurality of routes extending from one parking start position to a plurality of parking positions. Further, the parking support device comprises route selecting means. The route selecting means selects the route on which the vehicle runs out of the plurality of routes, during automatic parking.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present disclosure relates to a parking assistance device. [Background technology]

[0002] Patent Document 1 discloses a parking assistance device that, in automatic parking with learning drives (hereinafter also referred to as learning-type automatic parking), when learning drives are repeated in nearby positions, and if the parking positions are different, provides guidance to start the learning drive from a different parking start position.

[0003] Conventionally, maps used for automatic parking were managed with the starting position as an index. Therefore, in automatic parking technology, if a learning drive was repeated at the same starting position, it was considered as a learning drive being redone, and even if the parking positions in the learning drives were different, only the parking position from the last learning drive was registered, which was a problem.

[0004] The technology of Patent Document 1 is a technology related to a countermeasure to the above problem. According to the technology disclosed in Patent Document 1, when the parking position is different, the starting position is changed and learning driving is performed, thereby generating multiple maps with different starting positions. Therefore, the technology of Patent Document 1 makes it possible to register different parking positions on the map. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2022-155282 A Summary of the Invention [Problem to be solved by the invention]

[0006] Here, FIG. 23 is a diagram showing an example of automatic parking processing by a parking assistance device using the technology of Patent Document 1. In the example of FIG. 23, a parking position PA is linked to a start position PZ using the technology of Patent Document 1. The example shows a case where an occupant of a vehicle VA selects parking at a parking position PA by instructing the start of automatic parking at the start position PZ, and after the parking assistance device starts automatic parking of the vehicle 1, the occupant notices that another vehicle VB is parked at the parking position PA when the vehicle VA reaches a position PY. At this time, the vehicle VA is automatically parked according to a map for parking at the parking position PA, so if another vehicle VB is parked at the parking position PA, the automatic parking cannot be continued.

[0007] For example, if the parking assistance device also stores a map for parking at parking position PB, and the space at parking position PB is available, it is possible to return to a parking start position (not shown) for parking at parking position PB and restart automatic parking from the beginning, but this is too troublesome. For example, it would be nice to be able to change the map to be referred to to another map during automatic parking and continue automatic parking, but this is not possible with conventional parking assistance devices. Thus, there is room for improvement in the convenience of the technology of learning-type automatic parking.

[0008] An object of the present disclosure is to provide a parking assistance device that improves the convenience of learning-type automatic parking. [Means for solving the problem]

[0009] In order to solve the above problems, the parking assistance device according to the present disclosure includes a map generation means and a driving control means. The map generation means registers a map including a route from a parking start position to a parking position during learning driving. The driving control means causes the vehicle to drive autonomously based on the map during automatic parking. The map also has a plurality of routes from one parking start position to a plurality of parking positions. Furthermore, the parking assistance device includes a route selection means. The route selection means selects a route along which the vehicle will drive from the plurality of routes during automatic parking. Effect of the Invention

[0010] According to the present disclosure, the convenience of learning-type automatic parking can be improved. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram showing an example of learning-type automatic parking performed by a parking assistance device according to an embodiment. [Diagram 2] FIG. 2 is a diagram showing a vehicle to which the parking assistance ECU according to the embodiment can be applied. [Diagram 3] FIG. 3 is a block diagram illustrating an example of a hardware configuration of the parking assistance ECU according to the embodiment. [Figure 4] FIG. 4 is a block diagram showing an example of the configuration of the parking assistance system according to the embodiment. [Diagram 5] FIG. 5 is a diagram illustrating an example of a process for identifying the positions of feature points according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of a process for recognizing a space based on a distribution of parallax according to the embodiment. [Figure 7] FIG. 7 is an example of a graph showing a distribution of disparity according to the embodiment. [Figure 8] FIG. 8 is a diagram illustrating an example of a process for determining whether parking is possible according to the embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of a process for identifying a deviation when the vehicle according to the embodiment deviates from the parking path. [Figure 10] FIG. 10 is a diagram illustrating an example of space recognition in the proposal process according to the embodiment. [Figure 11] FIG. 11 is a diagram showing an example of an image proposing an additional parking position in the proposal process according to the embodiment. [Figure 12] FIG. 12 is a diagram illustrating an example of the second analysis process according to the embodiment. [Figure 13] FIG. 13 is a diagram illustrating an example of the second analysis process according to the embodiment. [Figure 14] FIG. 14 is a diagram illustrating an example of a structure of route data according to the embodiment. [Figure 15] FIG. 15 is a diagram illustrating another example of the structure of the route data according to the embodiment. [Figure 16] FIG. 16 is a diagram illustrating an example of a feature point registration process according to the embodiment. [Figure 17] FIG. 17 is a diagram illustrating an example of a feature point registration process according to the embodiment. [Figure 18] FIG. 18 is a diagram illustrating an example of a feature point registration process according to the embodiment. [Figure 19] FIG. 19 is a diagram illustrating an example of the continuous learning traveling according to the embodiment. [Figure 20] FIG. 20 is a diagram illustrating an example of the continuous learning traveling according to the embodiment. [Figure 21] FIG. 21 is a flowchart illustrating an example of a process executed by the parking assistance ECU according to the embodiment. [Figure 22] FIG. 22 is a diagram illustrating an example of automatic parking executed by the parking assistance ECU according to the embodiment. [Diagram 23] FIG. 23 is a diagram for explaining an example of an automatic parking process of a parking assistance device using the technology of Patent Document 1. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. Note that each embodiment described below shows a specific example of the present disclosure. Therefore, each component, the arrangement position and connection form of each component, each step and the order of each step shown in the following embodiments are merely examples and are not intended to limit the present disclosure. In addition, among the components in the following embodiments, components that are not described in the independent claims will be described as optional components.

[0013] In addition, each drawing is a schematic diagram and is not necessarily a precise illustration. In each drawing, the same reference numerals are used for substantially the same configurations, and duplicated explanations are omitted or simplified.

[0014] (Learning-based automatic parking) Fig. 1 is a diagram showing an example of the learning type automatic parking according to this embodiment. In the learning type automatic parking, the vehicle is manually parked in advance and the starting position and parking route are stored. In the learning type automatic parking, when the driver activates the automatic parking function at the stored starting position, the vehicle 1 can automatically drive and park along the parking route that was used when the vehicle was manually parked in advance.

[0015] More specifically, in learning-type automatic parking, the vehicle 1 generates a map including the position information of features while driving for manual parking (hereinafter also referred to as learning driving), and when driving for automatic parking, the vehicle 1 uses the map to estimate the position and attitude of the vehicle 1 while driving.

[0016] Here, the term "geographical features" refers to stationary objects around the vehicle, such as road markings and obstacles. The position of the geographical feature may be determined by determining the direction of the object from the positions of the object's images captured by multiple cameras, and then specifying the coordinates based on the principle of triangulation. Alternatively, the coordinates may be determined by using a distance measuring device such as sonar or radar, and based on the distance from multiple positions and the principle of trilateration.

[0017] The map for adaptive automatic parking includes location information of features (hereinafter also referred to as feature information) and parking route data. For example, as shown in FIG. 1, the parking route is a route connecting a parking start position 15 and a parking position 18. The parking route may be divided into straight sections and curved sections by dividing the route into end points 16 and 17, which are points where the steering angle changes. The parking route data may also be registered as a collection of data for multiple sections. The data for each section may be accompanied by information on the length, and the data for the curved section may be accompanied by information on the steering angle and turning radius. The parking route data according to this embodiment is made up of data for multiple sections.

[0018] The feature information may be, for example, data of a plurality of feature points extracted from a camera image. A feature point is a point whose position can be specified by extracting the image of a feature captured in a camera image, and the data of the feature point to be registered in a map is accompanied by position information. The position information may be position information of a feature specified by the principle of trilateration, or position information specified by the principle of triangulation described above. Since existing methods can be used to extract feature points, a detailed description will be omitted.

[0019] During learning driving, the parking assistance device generates a map including feature information (data on multiple feature points) and a parking route (data on multiple sections). During automatic parking, the parking assistance device estimates the position and attitude of the vehicle by comparing the data on multiple feature points read from the map with the data on multiple feature points extracted from a camera image. The parking assistance device then controls the vehicle so that the position of the vehicle follows the multiple sections included in the parking route read from the map in order.

[0020] (Vehicle configuration) Fig. 2 is a diagram showing a vehicle 1 to which the parking assistance device in the embodiment can be applied. As shown in Fig. 2, the vehicle 1 is equipped with a parking assistance system 1S. The parking assistance system 1S includes an operation device 10, an HMI (Human Machine Interface) device 20, a vehicle control device 30, a navigation device 40, a sonar ECU 50, and a parking assistance ECU 100.

[0021] Other devices may be mounted on the vehicle 1. In addition, in Fig. 2, the operation device 10, the HMI device 20, the vehicle control device 30, the navigation device 40, the sonar ECU 50, and the parking assistance ECU 100 are illustrated as separate devices, but some or all of these devices may be integrated together.

[0022] The operation device 10, the HMI device 20, the vehicle control device 30, the navigation device 40, the sonar ECU 50, and the parking assistance ECU 100 will be described later.

[0023] Cameras 2a, 2b, 2c, and 2d are provided at four locations on the front, rear, left, and right sides of the body of vehicle 1. Hereinafter, when there is no particular distinction between cameras 2a, 2b, 2c, and 2d, they will be simply referred to as cameras 2. Each camera 2 is equipped with a fisheye lens and has a horizontal viewing range of 180 degrees or more (see dashed lines).

[0024] Since each camera 2 is mounted at a depression angle in order to capture the road surface, when the range of the road surface captured is converted into a horizontal field of view, a range of about 240 degrees of the road surface is captured by each camera 2. For example, the front and rear wheels and the side of the vehicle 1 body are captured in the images captured by the side cameras 2a and 2b installed on the left and right sides of the vehicle 1 body.

[0025] The locations and number of cameras 2 are not limited to the example shown in FIG.

[0026] 2, twelve sonar sensors 3a to 3l are installed in the vehicle 1. Hereinafter, when there is no particular distinction between the sonar sensors 3a to 3l, they are simply referred to as sonar sensors 3. For example, on the left side of the vehicle 1, sonar sensor 3a is installed on the front side (FLS: Front Left Side) of the vehicle 1, and sonar sensor 3b is installed on the rear side (BLS: Back Left Side) of the vehicle 1.

[0027] In addition, on the right side of the vehicle 1, a sonar sensor 3c is installed on the front side (FRS: Front Right Side) of the vehicle 1, and a sonar sensor 3d is installed on the rear side (BRS: Back Right Side) of the vehicle 1. These four sonar sensors are also called side sonars because they detect obstacles on the sides of the vehicle.

[0028] In addition, in front of the vehicle 1, sonar sensor 3e (FLC: Front Left Corner), sonar sensor 3f (FL: Front Left), sonar sensor 3g (FR: Front Right), and sonar sensor 3h (FRC: Front Right Corner) are installed in that order from the left side in the forward direction of the vehicle 1.

[0029] The sonar sensors 3f and 3g provided on the inside detect obstacles in the traveling direction when the vehicle 1 travels straight. The sonar sensors 3e and 13h provided on the outside detect obstacles in the turning direction when the vehicle 1 turns. The sonar sensors 3e and 3h are also called corner sonars. The detection ranges of the four sonar sensors 3e, 3f, 3g, and 3h are shown as triangles in FIG. 1, but the detection ranges are not limited to the triangular range and can detect up to about 10 m from the vehicle. The detection ranges of adjacent sonar sensors are installed so as to overlap each other.

[0030] In addition, at the rear of the vehicle 1, sonar sensor 3i (BLC: Back Left Corner), sonar sensor 3j (BL: Back Left), sonar sensor 3k (BR: Back Right), and sonar sensor 3l (BRC: Back Right Corner) are installed, in that order from the left side in the forward direction of the vehicle 1.

[0031] The sonar sensors 3j and 3k provided on the inside detect obstacles in the direction of travel when the vehicle 1 reverses. The sonar sensors 3i and 3l provided on the outside detect obstacles in the direction of the turn when the vehicle 1 reverses and turns. The sonar sensors 3i and 3l are also called corner sonars. In FIG. 2, the directions in which the four sonar sensors 3i, 3j, 3k, and 3l detect and their fan-shaped spread are illustrated as triangles, but the detection ranges are not limited to the inside of the triangles in the figure and extend beyond them. For example, adjacent sonars are installed so that their detection ranges overlap each other. The same is true for the four sonar sensors 3e, 3f, 3g, and 3h on the front side of the vehicle.

[0032] The detection ranges of the sonar sensors 3a, 3b, 3c, and 3d installed on the sides of the vehicle 1 are set to be narrower than the detection ranges of the sonar sensors installed in front and rear of the vehicle 1. This is to improve the position resolution when the parking assist ECU 100 detects parking spaces on the sides of the vehicle 1 by minimizing overlap of the side sonar detection ranges when the vehicle 1 moves.

[0033] Furthermore, each sonar sensor is installed at a height and depression angle that allows easy detection of surrounding obstacles when parking vehicle 1. Note that the installation locations and number of sonar sensors 3a to 3l are not limited to the example shown in FIG.

[0034] Here, in this embodiment, sonar refers to a sonar system made up of the above-mentioned sonar sensors 3a to 3l and the sonar ECU 50. The sonar ECU 50 is a control device that performs overall control of the sonar system.

[0035] The sonar sensor 3 emits directional sound waves and receives the reflected waves. The sonar ECU 50 detects the distance to an obstacle based on the time from when the sonar sensor 3 emits sound waves to when it receives the reflected waves. The sonar ECU 50 detects the surroundings of the vehicle using multiple sonars, and identifies the position of the obstacle based on the distance from the multiple sonar sensors 3 that detect it.

[0036] The sonar system detects an obstacle in the traveling direction of the vehicle 1 using a sonar sensor 3 provided on the bumper of the vehicle 1. When the sonar ECU 50 detects an obstacle, it determines whether the vehicle 1 will collide with the obstacle. When the sonar ECU 50 determines that the vehicle 1 will collide with the obstacle within a predetermined time, it instructs the vehicle control device 30 to activate the automatic brake.

[0037] This ensures that the vehicle 1 does not collide with obstacles in front or behind it even during automatic parking. In addition, side sonar installed on the left and right sides of the vehicle 1 can detect when the side of the vehicle approaches an obstacle due to the difference in inner wheel radius.

[0038] Note that sonar is not an essential component of the parking assistance system 1S. In other words, the vehicle 1 may be configured not to include sonar. In this case, for example, the parking assistance ECU 100 may detect the distance to an obstacle by processing a camera image captured by the camera 2. For example, the parking assistance ECU 100 may determine whether the vehicle 1 will collide with the obstacle within a predetermined time. If it is determined that a collision will occur, the parking assistance ECU 100 may instruct the vehicle control device 30 to activate an automatic brake.

[0039] (Parking Assist ECU Hardware Configuration) Next, a hardware configuration of the parking assist ECU 100 will be described. The parking assist ECU 100 is an example of a parking assist device. Fig. 3 is a diagram showing an example of a hardware configuration of the parking assist ECU 100 according to an embodiment. The functions of the parking assist ECU 100 described later may be implemented in the hardware shown in Fig. 3.

[0040] The parking assistance ECU 100 may be a computer including a CPU 101, a ROM 102, a RAM 103, an I / O (input / output interface) 104, an IMP (Image Processor) 105, and a communication I / F (Interface) 106, with each element being connected via a bus.

[0041] The parking assistance ECU 100 may accommodate multiple elements on a single chip. Also, one element may be configured with multiple chips in the parking assistance ECU 100. A single bus is not required, and multiple types of buses may be combined.

[0042] For example, the CPU 101, ROM 102, RAM 103, IMP 105, and communication I / F 106 may be accommodated in one chip and connected by a parallel bus, while the I / O 104 may be composed of multiple chips and connected to the chip accommodating the CPU 101 via a serial bus.

[0043] The parking assistance ECU 100 performs automatic parking by obtaining information from other devices (e.g., the navigation device 40) and giving instructions to other devices (e.g., the vehicle control device 30) via the communication I / F and the in-vehicle LAN.

[0044] The CPU 101 controls the entire parking assist ECU 100. The functions of each part of the parking assist ECU 100 may be implemented in the form of a program executed by the CPU 101. The ROM 102 and RAM 103 correspond to storage parts, and the ROM 102 corresponds to a non-volatile area. The RAM 103 is used for temporary storage as a working area for the CPU 101. For example, the RAM 103 temporarily stores camera images such as display images and detection images, information on detected feature points, and the like.

[0045] The IMP 105 is a processor that is specialized for image processing and parallel processing and has improved processing performance. Some of the functions of the parking assistance ECU 100 (e.g., the image processing unit 130, the space recognition unit 140, the position estimation unit 150, etc.) described later may be executed by the IMP 105.

[0046] (Parking Assist ECU function) Next, the function of the parking assistance ECU 100 will be described. Fig. 4 is a block diagram showing an example of the configuration of the parking assistance system 1S according to the embodiment. As shown in Fig. 4, the cameras 2 (2c(F), 2d(B), 2a(L), 2b(R)) output the captured camera images. The image processing unit 130 of the parking assistance ECU 100 receives the camera images and performs the generation of a display image, etc. The display image is output from the notification unit 180 to the HMI device 20.

[0047] The notification unit 180 superimposes a message on a display image or outputs a voice message in response to an instruction from the state management unit 110. The notification unit 180 is an example of a notification means, and the HMI device 20 to which the message is output and the state management unit 110 that instructs the output of the message may also be included in the notification means. In other words, the notification means is a functional element that conveys information to the occupant. Since the state management unit 110, the notification unit 180, and the HMI device 20 are involved in the notification, these are sometimes referred to as the HMI device 20, etc.

[0048] The state management unit 110 receives a user's operation via the operation device 10, and controls the functions of the parking assistance ECU 100 in response to the user's operation.

[0049] The state management unit 110 receives position information from a main body (not shown) of the navigation device 40. When performing a learning drive, the state management unit 110 adds position information of the parking start position to a map and records the map in the storage unit 170. When performing automatic parking, the state management unit 110 compares the position information of the navigation device 40 with the position information added to the map, and selects a usable map.

[0050] The image processing unit 130 generates a display image and a detection image. The detection image is, for example, an image in which changes in brightness and color, that is, contrast, are emphasized. The image processing unit 130 extracts feature points from the detection image. The image processing unit 130 extracts points that are corners of an image or ends of lines that can be specified as points, rather than surfaces or sides of an image, as feature points. The information on the feature points output by the image processing unit 130 includes information on the color and shape of the image, and position information on the feature points on the camera image.

[0051] The space recognition unit 140 identifies the positions and distribution of the characteristic points. In addition, the space recognition unit 140 identifies characteristic points that are not on the road surface, i.e., obstacles, by analyzing the positions and distribution of the characteristic points. Then, the space recognition unit 140 recognizes an area without obstacles, i.e., space, based on the positions and distribution of the identified obstacles.

[0052] FIG. 5 is a diagram for explaining an example of a process in which the space recognition unit 140 specifies the position of a feature point. As shown in FIG. 5, the space recognition unit 140 specifies the position of the feature point (subject) P captured by the camera 2 of the vehicle 1 by motion parallax. First, the space recognition unit 140 converts the position of the feature point P on the camera image into the angle (azimuth angle) of the feature point P with respect to the vehicle 1. For example, assume that the camera 2 of the vehicle 1 moves from point A to point B on the Y axis as the vehicle 1 moves. Then, the image of the feature point moves on the camera image, and the azimuth angle found from the position of the feature point P captured by the camera 2 changes from θ1 to θ2. Such a change in azimuth angle (θ1→θ2) is called motion parallax.

[0053] The distance between points A and B (the length of the line segment AB) is the amount of movement of the vehicle 1, and can be determined from the number of rotations of the wheels. The spatial recognition unit 140 can determine the XY coordinates (x, y) of the feature point P from the coordinates of the line segment AB, its length, θ1, and θ2, based on the principle of triangulation. The vertical position of the feature point P captured by the camera 2 in the image corresponds to the height of the feature point P. For example, if the feature point P is the tip of a stick standing vertically on the ground, it will be captured on the screen above the base of the stick. Therefore, if the spatial recognition unit 140 determines the XY coordinates, it can determine the Z coordinate based on the vertical position in the image. In other words, the spatial recognition unit 140 can determine the three-dimensional coordinates of the feature point (subject) P by applying the motion parallax that appears in camera images taken at different times to the principle of triangulation.

[0054] Also, the space recognition unit 140 recognizes the space based on the distribution of parallax. FIG. 6 is a diagram for explaining an example of a process for recognizing the space based on the distribution of parallax. Also, FIG. 7 is an example of a graph showing the distribution of parallax. For example, as shown in FIG. 6, a vertical wall 201 is standing in parallel with the path of the vehicle 1. Then, the position of the camera 2 is point A, the intersection point (directly below the camera 2) between the road surface and the perpendicular line from point A is point B, the intersection point between the wall 201 and the perpendicular line from point B is point C, and the intersection point between the wall 201 and the perpendicular line from point A is point D, and the road surface continues from the bottom of the vehicle 1 to point C. Point E in the diagram is a point on the perpendicular line BC. Also, points B, C, D, and E, and line segments BC and CD each have a characteristic point.

[0055] In this case, in Fig. 7, the horizontal axis indicates the angle θ indicating the direction of the feature point, with the angle directly below camera 2 being 0 degrees, and the vertical axis indicates the parallax occurring at the feature point. As shown in Fig. 7, the graph showing the parallax of the feature point captured in the angle θ direction starts to decrease and then increases at the angle θc of point C. The reason for this will be explained below.

[0056] The lower diagram of FIG. 6 shows the parallax caused by the movement of the vehicle at the feature points C and E located right beside the camera 2, where the motion parallax of the feature point C is θ3, and the motion parallax of the feature point E in the middle is θ4. In this example, the motion parallax θ4 of the nearby point B is larger than the motion parallax θ3 of the farther point C. From this, it can be seen that the closer to the camera 2, the larger the parallax. Therefore, between points B and C on the road surface (on the line segment BC, in the range d1 in FIG. 7), the larger the θ, the smaller the parallax. In contrast, between points C and D on the wall surface of the wall 201 (on the line segment CD, in the range d2 in FIG. 7), the smaller the difference in elevation with the camera 2, the shorter the distance, so the parallax becomes smaller as the distance becomes shorter. Therefore, in the example of FIG. 6, between points C and D on the wall surface of the wall 201, the larger the θ, the larger the parallax becomes. If the wall 201 were not present, the parallax would have continued to decrease as indicated by the dotted line in FIG. 7. Therefore, based on the fact that the parallax rises at θc, it can be determined that a three-dimensional object continues above point C.

[0057] In this way, the space recognition unit 140 can identify the position of a three-dimensional object such as a wall 201 by analyzing the distribution of parallax. Furthermore, since there is no discontinuity in the distribution of parallax up to point C (range d1), this indicates that a road surface without steps is visible. Furthermore, since there is no discontinuity in the distribution of parallax between points C and D (range d2), it is possible to identify the presence of a wall 201. Furthermore, it is possible to identify from the distribution curve that the wall 201 is vertical. In this way, the space recognition unit 140 can identify that there is a space to the side of the vehicle 1 from the distribution of parallax.

[0058] In addition, the space recognition unit 140 detects a parking space when approaching a parking position and determines whether parking is possible. Figure 8 is a diagram for explaining an example of a process for determining whether parking is possible. Here, a case is considered in which the space recognition unit 140 of the vehicle 1A detects a parking space and determines whether parking is possible in the situation shown in Figure 8.

[0059] In the example of FIG. 8, a triangular cone 202 is placed at the back of the parking space. In FIG. 8, a vehicle 1B is parked in the parking space, but the vehicle 1B was not present during the learning run. In other words, it is assumed that the feature points of the triangular cone 202 at the back and the feature points in the parking space were registered in the map during the learning run. In this case, the space recognition unit 140 can compare the observed motion parallax with the motion parallax during the learning run. For example, the motion parallax observed when the direction of the floor of the parking space is captured from the vehicle 1A is larger than when the vehicle 1B is not present and the floor of the parking space is captured in the camera image, because the distance to the subject is shortened due to the presence of the vehicle 1B.

[0060] For example, the spatial recognition unit 140 may analyze the distribution of motion parallax on the floor of a parking space, and if it determines that there is an area where the parallax is larger than during the learning drive, as described above, it may determine that parking is not possible because there is a three-dimensional object in the parking space.

[0061] Furthermore, if the feature points at the back of the parking space or on the floor of the parking space, which were registered during the learning drive, are not detected during automatic parking, the space recognition unit 140 may determine that parking is not possible because of an obstructing object. In the example of Fig. 8, the space recognition unit 140 may determine that parking is not possible if the feature points representing the triangular cone 202, which were registered during the learning drive, are not detected.

[0062] In this embodiment, an example will be described in which the space recognition unit 140 recognizes the space by detection using the camera 2. However, the space recognition unit 140 may determine whether parking is possible using an obstacle detection means such as sonar or radar. For example, when the vehicle 1 is parked, if the distance to the nearest obstacle on the side of the vehicle 1 is equal to or greater than the vehicle width of the vehicle 1, the space recognition unit 140 may estimate that there is a space available for parking on the side.

[0063] The space recognition unit 140 may also perform processing by combining detection using the camera 2 and obstacle detection using sonar or the like. The space recognition unit 140 processes feature points detected using the camera 2, while also making a comprehensive judgment using obstacle detection information obtained by sonar. For example, even if the feature points of a triangular cone 202 that was at the back of the parking space during learning driving are not captured in the camera image during automatic parking, it may be determined that parking is possible as long as the sonar does not detect any obstacles within a range corresponding to the floor surface of the parking space.

[0064] The position estimation unit 150 functions during automatic parking and estimates the position and attitude of the vehicle based on the feature points registered in the map and the feature points on the camera image. The map may also include a camera image taken at the start of the learning drive.

[0065] Specifically, when automatic parking begins, the position estimation unit 150 compares information on feature points detected by the image processing unit 130 (information on the color and shape of the image, and information on the positions of feature points on the camera image) with information on feature points registered on a map, and identifies feature points that match the map.

[0066] If the position of the vehicle 1 at the start of automatic parking is close to the position at which the learning run started and there is no significant difference in the orientation (posture) of the vehicle body of the vehicle 1, it can be expected that the positions of the feature points on the camera image at the start of automatic parking will not be significantly different from the positions of the feature points on the camera image at the start of learning run. Therefore, the position estimation unit 150 may extract feature points that match the map by comparing the positions of the feature points on the camera image at the start of automatic parking and the start of learning run. Alternatively, the corresponding feature points may be identified by a comprehensive matching process that compares the arrangement of the detected feature points with the arrangement of the feature points registered on the map, and also compares the color and shape information of the image. Here, the matching process is a process of identifying feature points on the camera image that match the feature points on the map.

[0067] After identifying the corresponding feature point, the position estimation unit 150 identifies the position and attitude (direction) of the vehicle at the start of automatic parking by comparing the orientation of the feature point registered in the map based on the starting position of the learning drive with the orientation of the feature point on the camera image. For example, since the left-right position of the feature point on the camera image corresponds to the orientation of the feature point relative to the vehicle, the position estimation unit 150 can identify the attitude of the vehicle based on the left-right position of the feature point.

[0068] In addition, since the map includes the three-dimensional coordinates of the feature points, and the vertical positions of the feature points on the camera image correspond to the depression angle, the position estimation unit 150 can identify the distance to the feature points registered on the map based on the vertical positions of the feature points. The position estimation unit 150 repeats these processes for the corresponding feature points, and identifies the most likely values ​​(maximum likelihood values) as the position and attitude of the vehicle.

[0069] As described above, the position estimation unit 150 identifies feature points on the map that match feature points on the image, and estimates the position and attitude of the vehicle 1 from the three-dimensional coordinates of the feature points on the map and the positions of the feature points on the camera image that match them. This processing is called self-position estimation. The self-position estimation processing and matching processing may be performed using existing methods, and detailed explanations will be omitted.

[0070] During learning driving, the driving control unit 160 periodically communicates with the vehicle control device 30 to acquire information on the number of rotations of the wheels and the steering angle of the vehicle 1. Then, based on the acquired information, it calculates the amount of movement and the amount of change in attitude of the vehicle 1 per unit time (the amount of change in the orientation of the body of the vehicle 1).

[0071] The traveling control unit 160 acquires the attitude of the body of the vehicle 1 by integrating the amount of change in attitude. The traveling control unit 160 acquires a movement vector of the vehicle 1 for each unit time from the acquired attitude and movement amount of the body of the vehicle 1. The traveling control unit 160 acquires the coordinates of the body at each time by integrating the acquired movement vectors. The traveling control unit 160 acquires a traveling trajectory diagrammed with broken lines by connecting the coordinates of the body at each time.

[0072] The above process is called vehicle position estimation by the driving control unit 160. Also, a route connecting the vehicle positions estimated by the vehicle position estimation is called a parking route based on the vehicle position estimation.

[0073] During learning driving, the driving control unit 160 may send the number of rotations and steering angle of the wheels of the vehicle 1, or the amount of movement and direction of movement (movement vector) of the vehicle 1, as parking route information to the map generation unit 120 from moment to moment, or may send the estimated vehicle position from moment to moment to the map generation unit 120. Note that the driving control unit 160 of this embodiment sends the parking route based on the vehicle position estimation to the map generation unit 120 at the end of learning driving. This parking route is a series of movement vectors for each unit time, and is therefore graphically represented in the form of broken lines from a bird's-eye view.

[0074] When automatically parking, the driving control unit 160 estimates the vehicle position using the above method and at the same time outputs an instruction value to the vehicle control device 30 so as to reproduce the steering angle and travel distance recorded on the map. The vehicle control device 30 controls the steering angle and vehicle speed according to the instruction value. However, the steering angle and vehicle speed follow the change in the instruction value with a delay and may temporarily deviate from the instruction value. In addition, the actual steering angle and vehicle speed follow the instruction value with an offset and may steadily deviate. As a result, the route estimated by the driving control unit 160 for the vehicle position may deviate from the parking route recorded on the map.

[0075] In such a case, the driving control unit 160 applies feedback control in a direction to pull the vehicle 1 back to the parking path. Specifically, the driving control unit 160 first controls the steering angle so that the path of the vehicle 1 intersects with the parking path, and when the vehicle 1 overlaps the parking path, controls the steering angle so that the path of the vehicle 1 follows the parking path. In other words, the driving control unit 160 estimates the position and attitude of the vehicle 1, and feedback controls the steering angle so that the path of the vehicle 1 follows the parking path recorded on the map.

[0076] The driving control unit 160 may reproduce the vehicle speed during learning driving when automatically parking. Also, the driving control unit 160 may limit the vehicle speed to a predetermined value or less when automatically parking. This is because if the vehicle speed is high, the wheels of the vehicle 1 may slip and deviate from the route. For example, the driving control unit 160 may maintain the vehicle speed during automatic parking at 5 km / h.

[0077] Since the above-mentioned position estimation unit 150 estimates its own position during automatic parking, it is preferable that the self-position estimation by the position estimation unit 150 be operated so as to be complementary to the vehicle position estimation by the driving control unit 160.

[0078] For example, in a turning section, the vehicle may deviate from the route due to slippage or centrifugal force, but the vehicle position estimation by the cruise control unit 160 has difficulty detecting deviations that are not reflected in the steering angle or wheel rotation speed.

[0079] Fig. 9 is a diagram for explaining an example of a process for identifying a deviation when the vehicle 1 deviates from the parking path. As shown in Fig. 9, for example, when the vehicle 1 is traveling on the arc AB and there are characteristic points at points A and B, the difference between the orientation of point A and the orientation of point B, that is, the magnitude of angle APB, is constant according to the circular angle theorem if the vehicle 1 is at point P on the arc APB.

[0080] However, when vehicle 1 is at point Q on arc AQB, which is outside of arc APB, the difference between the orientation of point A and the orientation of point B, i.e., the magnitude of angle AQB, is smaller than angle APB. Also, the angular difference between angles AQB and APB corresponds to the magnitude of the deviation. In this way, by capturing characteristic points in front of and behind vehicle 1 with camera 2 and evaluating their orientations, the direction and magnitude of deviation can be identified when vehicle 1 deviates from the parking path.

[0081] Therefore, for example, when estimating that slippage has occurred from the number of rotations of the wheels of the vehicle 1 or in a turning section with a small turning radius, the driving control unit 160 may evaluate that the reliability of its own vehicle position estimation is low and use the estimated value of the position estimation unit 150. In this case, the driving control unit 160 may obtain the self-position estimation data (position and attitude) of the position estimation unit 150, overwrite the self-position estimation data of its own vehicle position estimation with the self-position estimation data, and ensure the reliability of subsequent vehicle control.

[0082] That is, the driving control unit 160 may back up its own vehicle position estimation with the self-position estimation by the position estimation unit 150. Furthermore, the driving control unit 160 may not use the data of its own vehicle position estimation during automatic parking, but may obtain the data of the self-position estimation by the position estimation unit 150 and perform feedback control of the vehicle 1 using only the data of the self-position estimation.

[0083] Returning to FIG. 4, the parking assistance ECU 100 will be described further. During learning travel, the map generation unit 120 stores the feature points, whose coordinates have been specified by the space recognition unit 140, and the coordinates of the feature points as part of the map data in the memory unit 170. This process of storing the feature points as part of the map data in the memory unit 170 is called "registering the feature points in the map." The map generation unit 120 is one of the map generation means. Since the map is generated by the memory unit 170 storing various data, it can be said that the memory unit 170, in addition to the map generation unit 120, is also included in the map generation means.

[0084] The map generating unit 120 also registers the parking path in the map. The parking path may be, for example, a parking path calculated by the driving control unit 160 based on the observed values ​​of the wheel rotation speed and steering angle of the vehicle 1 and estimated by the vehicle's position. In other words, the map generating unit 120 may register the parking path generated by the driving control unit 160 and illustrated in the form of broken lines from a bird's-eye view in the map as it is.

[0085] The map generating unit 120 of this embodiment receives the parking route, which is diagrammed in the form of a broken line from a bird's-eye view, from the driving control unit 160 at the end of the learning drive, and reconstructs it into multiple sections connecting the parking start position and the parking position. Each section is either a straight section where the vehicle travels straight with a zero steering angle, or a turning section where the vehicle turns with a constant steering angle, and is an approximation of the broken line parking route. This can be said to be the generation of a parking route that approximates the route of the learning drive. Note that for sections whose end points are the parking start position or the parking position, the approximation is performed under the constraint that the parking start position or the parking position must not be changed.

[0086] Then, the map generating unit 120 registers the distance traveled in each section and the steering angle in the turning section in the map. In this way, by simplifying the broken line route into a small number of sections and registering them in the map, the steering angle control and vehicle speed control during automatic parking become easier and the amount of map data becomes smaller. In addition, the vehicle body sways less during automatic parking, which increases the passenger's favorability.

[0087] In other words, the map generator 120 of the parking assist ECU 100 according to this embodiment does not completely reproduce the parking route during learning travel for the automatic parking route, but reproduces the parking position during learning travel for the parking position. This is because, in general, the occupant does not evaluate the position of the vehicle 1 on the parking route very much, but evaluates the accuracy of the parking position.

[0088] In addition, since the positions of objects around the parking route may change, even if the parking route during learning is perfectly reproduced, it is not guaranteed that the vehicle 1 will not collide with an obstacle. For this reason, in this embodiment, the parking assistance ECU 100 uses an automatic brake system to ensure that the vehicle 1 does not collide with an obstacle. In other words, reproducing the parking route during learning does not produce any particular effect, so it can be said that there is less need for it compared to reproducing the parking position.

[0089] In this embodiment, the map generating unit 120 is assumed to register the feature points around the parking route and generate the parking route, but the parking route may be generated by another functional unit. For example, the parking assist ECU 100 may include a route generating unit as a functional unit in addition to the map generating unit 120.

[0090] In this case, the map generating unit 120 may register the characteristic points around the parking route and the parking route generated by the route generating unit. Alternatively, the route generating unit may generate the parking route and register the generated parking route on the map. In other words, the map generating unit 120 and the route generating unit may cooperate to generate the map.

[0091] (Map creation process) Next, the map generation process will be described. As a term, manual driving in which the vehicle is driven manually from the parking start position to the parking position and the parking start position and the parking position are stored will be called complete learning driving. Complete learning driving is the same as what has been conventionally called learning driving. On the other hand, in the learning-type automatic parking method according to this embodiment, there are multiple types of learning driving that do not pass through the parking start position. Therefore, in order to distinguish them, learning driving in which the vehicle is driven from the parking start position to the parking position will be called complete learning driving. Below, the map generation process will be described using complete learning driving as an example.

[0092] In this embodiment, the process of generating a map by a complete learning drive can be divided into the following processes (sub-processes).

[0093] P001: Starting point processing (registering the GPS coordinates of the starting position) P002: Collection and processing (collection of information on characteristic points, route information such as steering angle and movement amount, detection images, etc.) P003: End point processing (accepting the completion of the learning run and notifying that analysis will be performed) P004: Analysis processing (arrangement of parking routes) P005: Proposal process (detection and presentation of parking location candidates) P006: Registration process (parking location and route registration)

[0094] The above map generation process is managed as a whole by the state management unit 110, but the occupant also participates in the process. Each functional unit shown in FIG. 4 performs a corresponding process under the management of the state management unit 110. Also, the occupant may perform a learning drive multiple times during the map generation process. Therefore, each functional unit may execute a corresponding process multiple times. Note that each functional unit does not necessarily execute all processes in response to a learning drive. For example, in a learning drive that does not start from a parking start position but adds a parking position, the map generation unit 120 does not execute the initial starting point process. Each process will be described in detail below.

[0095] (P001: Start point processing) The start point process is a process performed at the start point of the learning drive (parking start position). The map generation unit 120 acquires the GPS coordinates of the start position from the navigation device 40 and registers them in the map so that the map to be used can be specified from the GPS coordinates at the start position (start point) of the learning drive during automatic parking. The map generation unit 120 also performs necessary initialization processes so that it can extract feature points, specify coordinates, and collect route information.

[0096] (P002: Collection process) The collection process is a process performed on the route between the start position and the parking position. The map generation unit 120 extracts characteristic points and identifies their coordinates, and collects information on the characteristic points (including coordinates). The map generation unit 120 also collects route information such as steering angle, vehicle speed, gear position, moving direction, and moving distance. The map generation unit 120 also saves detection images and detection data on the route and at the parking position.

[0097] (P003: End point processing) The end point processing is a process performed at the end point (parking position) of the learning drive. The state management unit 110 starts the end point processing when the gear position becomes P. The state management unit 110 notifies the occupant to wait while the data is analyzed. The map generation unit 120 starts the next analysis process after the above notification by the state management unit 110. Note that the state management unit 110 may cause the map generation unit 120 to start the analysis process while outputting the notification message. In this way, the state management unit 110 can shorten the waiting time felt by the occupant.

[0098] In this map generation process, the processes P001 to P003 are processes during learning driving. Moreover, the processes from P004 are processes after learning driving. The map generation unit 120 may avoid processes with a high load during learning driving and prioritize data collection, and perform the processes with a high load after learning driving.

[0099] For example, in the collection process P002, the map generation unit 120 may only store the detection image used for spatial recognition, and the spatial recognition itself may be performed by the spatial recognition unit 140 during the analysis process. In this embodiment, the map generation unit 120 is configured to only collect information on characteristic points in the collection process, and select the characteristic points and register the information on the characteristic points after the learning drive. Alternatively, the map generation unit 120 may register the information on characteristic points during the collection process, and delete unnecessary information on characteristic points in the analysis process after the learning drive to reduce the amount of data.

[0100] (P004: Analysis processing) The analysis process is mainly performed by the map generation unit 120. During the end point process, the map generation unit 120 receives the parking route that is diagrammed by the driving control unit 160 using broken lines. During the analysis process, the map generation unit 120 approximates the received parking route with a straight section where the vehicle travels straight with a zero steering angle and a turning section where the vehicle turns with a constant steering angle.

[0101] Then, a large amount of data describing many broken lines is replaced with a small amount of data describing a small number of sections. As a result, the route information is consolidated into the distance traveled in each section and the steering angle of the turning section. The start and end points of each section are called endpoints. Once the coordinates of the endpoints of each section are determined, the distance traveled in each section is determined. Therefore, instead of omitting the distance traveled in each section from the data, the coordinates of the endpoints of each section may be included in the data.

[0102] The above process is a route calculation process that sets a parking route from the start position of the learning drive to the parking position so as to generally follow the route of the learning drive. Therefore, existing route calculation methods can be applied to the above process. For example, the map generation unit 120 may set multiple end points on the route of the learning drive and calculate a parking route consisting of a straight section and a turning section that extends from the parking start position to the parking position via the multiple end points.

[0103] The process of simplifying the parking route is not essential. In other words, the map generator 120 may register the parking route that the driving control unit 160 has drawn with broken lines as the parking route. In this case, the route calculation process of setting a route that generally follows the learned driving route is not necessary.

[0104] The map generating unit 120 may perform the analysis process only once for one learning run, but may also perform the analysis process multiple times. For example, when a parking position different from the parking position of the learning run is added in the next proposal process, the map generating unit 120 repeats the analysis process to calculate a parking route for parking at the added parking position. In other words, when a parking position different from the parking position is set after the learning run, the map generating means generates a different route from the parking start position of the learning run to the different parking position, and registers the generated different route in the map. Then, a map having multiple routes from one parking start position to multiple parking positions can be generated, so that a parking position can be selected according to the generated map.

[0105] (P005: Proposal processing) The proposal process is a process in which the state management unit 110 proposes to the occupant of the vehicle 1 to additionally register a parking position. For example, the map generation means (storage unit 170) may collect detection information around the vehicle during learning driving, and set a different parking position different from the parking position of the learning driving based on the detection information. Then, a different route from the route of automatic parking to the different parking position is generated, and the generated different route is registered in the map. In this way, multiple parking positions and parking routes can be registered in one learning driving, so that the occupant's work can be reduced. During the proposal process, the state management unit 110 may propose additional registration when the space recognition unit 140 detects a space where parking is possible. In addition, the state management unit 110 may propose additional registration regardless of detection by the space recognition unit 140. Hereinafter, the setting of another parking position by detection will be called automatic setting, and the setting of another parking position not by detection will be called manual setting. Hereinafter, the former automatic setting will be described, and the latter manual setting will be described later.

[0106] The space recognition unit 140 may detect a space available for parking at the parking position. The space recognition unit 140 may also detect a space available for parking along the parking path. In other words, the detection for automatic setting may be performed during the collection process, during the proposal process, or across both periods. For example, if detection images and detection data are saved during the above-mentioned collection process, it becomes possible to detect a space available for parking along the parking path during the proposal process. Furthermore, processing of the saved data may be included in the above-mentioned analysis process.

[0107] 10 is a diagram for explaining an example of space recognition in the proposed process. For example, the sonar system detects the distance to a lateral obstacle (wall surface) using the side sonars 3a and 3b when the vehicle 1 is moving straight from the parking start position 19 to the turning position 21. The space recognition unit 140 may analyze the detection result, and estimate that there is a space available for parking on the side of the vehicle 1 when a space without obstacles up to a distance 1.4 times the vehicle width of the vehicle 1 continues for a length 1.2 times the vehicle length of the vehicle 1.

[0108] 10, after the vehicle 1 is parked at the parking position 22, the space recognition unit 140 may estimate that there is a space available for parking between the parking position and the wall on the left side by the same method as described above. Furthermore, the space recognition unit 140 may detect motion parallax from an image stored for detection without relying on a side sonar, and estimate that there is a space available for parking by identifying the range in which the road surface is visible by analyzing the distribution of parallax.

[0109] When the second parking position (a parking position different from the parking position) is automatically set by the detection, as described later, an additional parking position may be proposed and registered on the map with the approval of the occupant, or may be automatically registered without approval. Alternatively, a passive notification may be given to obtain tacit approval. For example, when it is determined that there is parking space for two cars in the garage during the learning drive, the second parking position may be automatically set without notifying the occupant. Alternatively, a message on the screen may be displayed to inform the occupant that the second parking position has been registered together with the first parking position, and the registration will be maintained unless the occupant performs an operation to cancel the registration. In this way, even if the first parking position is blocked during automatic parking, the vehicle can change course to the second parking position and continue automatic parking. The fact that the second parking position has been automatically set may be notified when parking in the first parking position is not possible, or after automatic parking has been performed.

[0110] FIG. 11 is a diagram showing an example of an image proposing an additional parking position in the proposal process. For example, when there is a space to the left of the parked vehicle 1 where the vehicle itself can fit, the state management unit 110 may display an image proposing an additional parking position as shown in FIG. 11. Specifically, the state management unit 110 may control the notification unit 180 to superimpose a semi-transparent image showing a virtual image of the vehicle parked adjacent to the overhead image output to the HMI device 20, and display a message such as "Do you want to register the space on the left as an additional parking position?" to inquire of the occupant whether to approve the request. Alternatively, the same message may be output by voice to request approval.

[0111] The image of vehicle 1 on the right side of Fig. 11 is a model of the vehicle itself. The remaining part is an overhead image generated by projectively transforming the camera image. For example, the notification unit 180 superimposes a model of the vehicle itself semi-transparently on the left side of the vehicle itself in the overhead image, allowing the viewer to imagine a scene in which two vehicles are parked in a garage.

[0112] The proposed additional parking position may be set based on the detection information. Specifically, the state management unit 110 may set the additional parking position by reflecting the distance between the vehicle 1 parked in the learning drive and a surrounding object (e.g., a wall). For example, as shown in FIG. 11, the state management unit 110 sets the additional parking position parallel to the parked vehicle 1, and sets the distance therebetween to the same as the distance between the parked vehicle 1 and the right wall 203. Furthermore, the distance from the additional parking position to the back wall is set to the same as the distance from the vehicle 1 to the back wall. In this way, the state management unit 110 can uniquely determine the additional parking position by reflecting the distance between the vehicle 1 parked in the learning drive and the surrounding object in the additional parking position. Since the parking position of the vehicle 1 reflects the distance at which the occupant feels safe and the occupant's preference, the additional parking position becomes preferable for the occupant by reflecting the distance between the parking position of the vehicle 1 and the surrounding object in the additional parking position.

[0113] Also, for example, the state management unit 110 may display an image showing the automatically set additional parking position on the touch panel of the navigation device 40, and the occupant may move the image of the additional parking position by operating the touch panel. In this way, if the automatically set additional parking position is not preferable, the occupant can manually adjust the additional parking position to suit his / her preference.

[0114] In the above case, for example, if the state management unit 110 displays an image proposing an additional parking position and then accepts the occupant's pressing of an approval button, the map generation unit 120 registers the proposed additional parking position. Also, if the state management unit 110 accepts the occupant's operation to move the additional parking position and then accepts the occupant's pressing of the approval button, the map generation unit 120 registers the additional parking position changed by the occupant's operation. In the former case, the additional parking position to be registered is a position that was automatically set, but since the occupant manually approves it, it can also be said to be manually set.

[0115] In order to distinguish such a method of setting an additional parking position only by the proposal process from a method of setting an additional parking position by an additional learning drive, this method is referred to as manual setting of an additional parking position. The proposal process may also be called a proposal means. The proposal means proposes to the occupant to add a parking position when the vehicle is parked during the learning drive, and sets another parking position based on the occupant's operation or approval. Then, the map generation means generates another route to the other parking position and registers the generated other route in the map. In this way, manual setting makes it possible to register an additional parking position without actually parking the vehicle, and the setting can be manually changed to suit one's preferences, which is simple and preferable.

[0116] As described above, detection information about the vehicle's surroundings may be collected during the learning drive, and parking position candidates may be automatically set based on the detection information and proposed to the vehicle occupant. When the parking position candidate is proposed to the occupant, it is set to a generally good parking position, so the occupant may make minor corrections and accept it, or may accept it as is. In this way, by using automatic setting in combination with manual setting, it is possible to support the occupant in setting and reduce the effort of manual adjustment.

[0117] Up to this point, the manual setting has been described in which the automatically set additional parking position is manually adjusted and set, but the additional parking position may be set manually instead of automatically. In other words, the detection process for detecting a space where parking is possible may be omitted, on the premise that the suggestion means allows the occupant to set the additional parking position. For example, the state management unit 110 may simply inquire of the occupant by displaying, for example, "Do you want to register another parking position?" on the HMI device 20. If the answer to the inquiry is Yes, the state management unit 110 may set the additional parking position according to the operation of the occupant received by the HMI device 20.

[0118] In the case where the detection process is not performed, multiple parking position candidates based on the layout of a general parking lot may be presented to the occupant, and the occupant may select or operate the parking position. For example, if the occupant answers Yes to the inquiry, the HMI device 20 may display an overhead image in which parking spaces indicating additional parking position candidates are arranged on the front, rear, left and right sides of the image of the vehicle 1. Then, the occupant selects the direction in which the parking position is to be set on the overhead image. In addition, the state management unit 110 may not display parking spaces in the direction in which a nearby obstacle is detected. For example, in the arrangement shown in FIG. 11, parking spaces are not displayed on the right side and rear of the vehicle 1. When the occupant touches one of the displayed parking spaces, the other parking spaces disappear, and the touched parking space moves following the movement of the finger. The mechanism for having the occupant operate the position of the space and confirming it with the approval button may be the same as the above-mentioned method.

[0119] Here, if the addition of a parking position is approved in the proposal process, the execution order of the map generation process will be as follows: P001: Start point process → P002: Collection process → P003: End point process → P004: Analysis process → P005: Proposal process → P004: Analysis process → P005: Proposal process → P006: Registration process. This is because the second analysis process is executed to calculate the parking route to the added parking position.

[0120] The second analysis process will be described below. FIG. 12 and FIG. 13 are diagrams for explaining an example of the second analysis process. The route RA (hereinafter, also referred to as the parking route RA) in FIG. 12 is a route traveled in the learning drive. The learning drive starts at the parking start position 23 and ends at the parking position 27 (hereinafter, also referred to as the first parking position 27), and the first analysis process and the proposal process are performed after parking. For example, in the second analysis process, the map generating unit 120 generates a route RB (hereinafter, also referred to as the parking route RB) that branches off from the parking route RA of the learning drive and parks at an additional parking position 32 (hereinafter, also referred to as the second parking position 32). If the route referred to when generating the route is the reference route, it can be said that the map generating means generates another route (for example, the parking route RB) based on the reference route (for example, the parking route RA). The reference route may be a route that has already been set, or may be a route registered in the map. By using such an existing route, the time required for route calculation can be reduced.

[0121] Generally, the second parking position 32 is often set parallel to the first parking position 27. For this reason, the map generating unit 120 may use the parking route RA when generating the parking route RB, and may use a part of the parking route RA translated in parallel as the parking route RB.

[0122] As shown in FIG. 12, the parking route RA is made up of a straight section 23-24, a turning section 24-25, a turning section 25-26, and a straight section 26-27.

[0123] Since route data is a collection of data on the sections that make up a route, generating another route based on a referenced route means generating data on the sections of the other route based on data on the sections that make up the referenced route. Specifically, when generating another route based on a referenced route, the map generating means generates data on the sections of the other route by copying, processing, or sharing data on the referenced section. This makes it possible to shorten the time required to generate a route and reduce the amount of route data.

[0124] The route to park at the second parking position 32, which is generated in the second analysis process, is set as the parking route RB. The map generating unit 120 may translate the sections of the parking route RA by the distance between the first parking position 27 and the second parking position 32, and set the sections of the parking route RB as the sections of the parking route RB. For example, as shown in FIG. 13, the map generating unit 120 translates the sections 24-25, 25-26, and 26-27 of the parking route RA to the sections 28-29, 29-31, and 31-32 of the parking route RB. Furthermore, the map generating unit 120 sets the section between the end point 24 and the end point 28 as the straight section 24-28.

[0125] The data of the translated section inherits the data of the original section. That is, the map generating unit 120 sets the data of the section 28-29, the section 29-31, and the section 31-32 to be the same as the data of the section 24-25, the section 25-26, and the section 26-27, respectively. In other words, the data of the reference section is copied and set as the data of the section to be generated. In addition, the map generating unit 120 sets the distance between the first parking position 27 and the second parking position 32 as the data of the distance of the straight section 24-28, and sets the steering angle data to zero because it is a straight section. In this way, the map generating unit 120 completes the setting of the parking route RB that starts at the end point 23, passes through the end points 24, 28, 29, and 31, and parks at the second parking position 32. The parking route RB branches off from the parking route RA at the end point 24, so the end point 24 is called a branch point. Moreover, a route that starts at the terminal point 24, passes through the terminal points 28, 29, and 31, and ends at the terminal point 32 is called a branch route.

[0126] A branch road is a type of route. When a branch road is connected to a route registered in a map starting from the parking start position or a route that has already been set, another route from the parking start position to another parking position is formed. In other words, another route from the parking start position to another parking position may be registered by registering a branch road in a map. For example, in a registration process described later, the map generating unit 120 may register in the map a branch road (parking route RB) that starts from the terminal point 24 (branch point) and passes through 28, 29, and 31 to park at the second parking position 32. Then, it is possible to park at the second parking position 32 by branching off at the terminal point 24 (branch point) from the parking route RA that starts from the terminal point 23 (parking start position) and parks at the first parking position 27. Alternatively, the parking route RB may be set as a route starting from the terminal point 23 (parking start position) and added to the map as an independent route separate from the parking route RA. In the latter case, the data of the first straight section 23-24 of the parking path RB is generated by duplicating (copying) the data of the first straight section 23-24 of the parking path RA.

[0127] When adding a branch road or a parking route to a map, the map generating unit 120 adds a link to the parking route. Here, a link is an address or offset that specifies a jump destination when jumping the data read position. The link added to the parking route jumps to either data of another route stored in the same map as the map that stores the route, or data of another route stored in a map other than the map that stores the route. In other words, if a link is added to the parking route, when the state management unit 110 is sequentially reading the parking route data from the memory unit 170 and automatically parking the vehicle, the parking route and parking position can be changed by jumping the read address according to the link that indicates the top address of the data of another route or the data of the branch road.

[0128] In the former case, a link to jump to data of section 24-28 of parking route RB is added to the data of section 23-24 of parking route RA. In the latter case, the map generating unit 120 adds a link to parking route RA, and also adds a link to jump to data of section 24-25 of parking route RA to data of section 23-24 of parking route RB generated by copying data of section 23-24 of parking route RA. This is an example of copying and then processing data of the referenced section. If a link to the other route is added to both routes RA and RB, when the state management unit 110 reads data of the parking route from the memory unit 170, it will be possible to jump to data of the other parking route regardless of which parking route data is read first.

[0129] The link may be specified by an absolute address or a relative address.

[0130] For example, if data of parking route RA is registered in the map data first, and data of a branch road is added, the address of the data of section 23-24 of parking route RA is already determined. If the data of the branch road is added later, the top address of the data of section 24-28 of the branch road is determined at that point. Therefore, the map generating unit 120 can calculate the difference in the addresses (address offset) and set it as a link to the data of section 23-24 of parking route RA.

[0131] When adding a branch road to the map, the state management unit 110 always starts reading the route from the parking route RA when reading the route from the storage unit 170. Then, the state management unit 110 can select whether to jump to the data of the branch road or to continue reading the data of the parking route RA when reading the link.

[0132] Also, when the parking route RB is registered separately from the parking route RA instead of registering a branch road, the map generating unit 120 may register the parking route RB in an independent map that is different from the map in which the parking route RA is registered. In this way, if one map corresponds to one parking route (parking position), it becomes easier to manage the map. In addition, since all map data is on the ROM, it is also possible to set a link to jump to another map, and even if a jump is made to another map, no delay occurs in processing.

[0133] Next, the data structure of the route data will be described. Fig. 14 is a diagram showing an example of the structure of route data. Fig. 14 illustrates an example of the data structure of route RA and route RB. As shown in the figure, route data is a collection of section data.

[0134] The section data consists of three words. A word is a unit of data having one address. One word may be, for example, 16 bits or 32 bits. The route data is included in the map data. The map data is written in an electrically rewritable ROM (for example, the ROM 102 of the parking assistance ECU 100). The map data can be accessed in word units.

[0135] The first word of the section data, Ax, indicates the steering angle and direction of travel. The second word of the section data, Dx, indicates the distance traveled in the section. The third word of the section data is the link. If the link is 0, it indicates that it is the final section, if it is 1 (+1), it indicates that there is no branch and the next section continues, and if it is other than 0 or 1, it indicates the relative address (offset) of the branch destination.

[0136] For example, the third word L1 of section 23-24 of route RA is the offset to the first word of section 24-28 of route RB. Also, the third word L2 of section 23-24 of route RB is the offset to the first word of section 24-25 of route RA. Normally, the first word of the next section is at the next address, so the offset when there is no branching is +1. Therefore, a link that is 1 may be called an implicit link when there is no branching.

[0137] When the map generating unit 120 reads out a link Lx that is neither 0 nor 1 from the storage unit 170, it adds Lx to the address if there is a branch, and increments the address by +1 if there is no branch. That is, for links in sections with a branch, the implicit link (+1) for cases where there is no branch is omitted. Also, since link = 0 indicates the final section, it is sufficient to stop the vehicle after driving the distance specified by the second word Dx and end automatic parking.

[0138] In the example of FIG. 14, the map generating unit 120 generates data for route RB using data for route RA. First, the map generating unit 120 copies the first section of route RA to set it as the first section 23-24 of route RB. Next, the map generating unit 120 inserts a straight section 24-28 after section 23-24. The second word D5 of the straight section 24-28 is set based on the distance between the first parking position 27 and the second parking position 32 in FIG. 13. Then, the map generating unit 120 copies the data for sections 24-25, 25-26, and 26-27 of route RA, respectively, and pastes them as data for sections 28-29, 29-31, and 31-32 of route RB, respectively, to complete the data for route RB.

[0139] In this way, by using the data of route RA, the map generating unit 120 can generate route RB with a small amount of calculation. In the example of Fig. 14, the data of route RA and the data of route RB are independent, so if the target parking position is the end point 27, the data of route RA is read, and if the target parking position is the end point 32, the data of route RB is read. Regardless of which route you start reading, you can change the route midway.

[0140] The data structure of the route is not limited to the above. Fig. 15 is a diagram showing another example of the data structure of the route. For example, the map generating unit 120 may share part of the data of the section of the reference destination (route RA) to generate data of the section of another route (route RB). Sharing the data of the section can reduce the amount of data of the route.

[0141] The example in FIG. 15 corresponds to a configuration in which a branch road is added to the data of route RA. Therefore, the state management unit 110 starts reading data from the first section 23-24 of route RA, regardless of the target parking position. Then, when the vehicle 1 reaches the end of the section, the state management unit 110 judges whether or not to branch. If branching and causing the vehicle 1 to travel along route RB, it jumps to the data of section 24-28 according to the link of section 23-24 (offset=10). Under the control of the travel control unit 160, the vehicle 1 travels straight for the distance indicated by D5.

[0142] Then, the vehicle 1 reaches the position of the end point 28 in Fig. 13, and thereafter, it is sufficient to travel along a route that is a parallel shift of sections 24-25 to 26-27 of the route RA. Therefore, the state management unit 110 jumps the read address to the data of section 24-25 according to the link of section 24-28 (offset = -11). As shown in Fig. 14, the data of sections 28-29 to 31-32 of the route RB is the same as the data of sections 24-25 to 26-27 of the route RA. Therefore, when the address is jumped to the data of sections 24-25 to 26-27 of the route RA and the same data is read, the vehicle 1 travels along a congruent route at a position that is a parallel shift of the distance indicated by D5.

[0143] As described above, if the second parking position can be set parallel to the first parking position, the map generator 120 can reduce the amount of calculation required for route generation and the amount of data to be handled by using data on existing sections. Also, if the second parking position is not parallel to the first parking position and there is an angle difference, a turning section may be inserted as the first section of the branch road, and the angle difference may be compensated for in the turning section, making it possible to use data on the section of the first route.

[0144] In the above, an example has been shown in which the map generating unit 120 stores data for each section of the route in consecutive addresses. However, the map generating unit 120 may distribute and arrange the data for each section. In this case, the data for a section requires a link to the next section when there is no branch. Therefore, the data for a section may include a data field indicating the number of links.

[0145] The data field indicates the range assigned to one piece of data. The data field may be a portion of bits in one word. The data field may also span multiple words. The number of links may be increased to allow branching in three or more directions from one branch point. Multiple branch points may also be provided to configure a parking route that allows parking at three or more parking locations.

[0146] Next, the second proposal process will be described. After the second analysis process is completed, the state management unit 110 performs the second proposal process together with the notification unit 180. If there is a space available for parking in a location that has not been set as a parking location in the second proposal process, the state management unit 110 may display a message such as "Do you want to register another parking location?" on the HMI device 20 or the like to inquire of the occupant.

[0147] Furthermore, the state management unit 110 may simply inquire of the occupant by displaying on the HMI device 20 or the like "Do you want to register another parking position?" without having the spatial recognition unit 140 perform spatial recognition. If there is no other space in which to set another parking position or if a parking position cannot be added due to restrictions such as the map data capacity, the state management unit 110 may notify the occupant by displaying on the HMI device 20 or the like "Parking position registration is completed" without making an inquiry.

[0148] If the occupant answers Yes to the inquiry, the same process as when the occupant answers Yes in the first proposal process is repeated, so a detailed explanation will be omitted. If the occupant answers No, the state management unit 110 notifies the occupant by displaying "Adding the parking position is completed" on the HMI device 20 or the like, and proceeds to the registration process.

[0149] (P006: Registration process) In the registration process, the map generating unit 120 registers data of the characteristic points on the map together with data of the parking route. Hereinafter, registering data of the characteristic points on the map will be referred to as registering the characteristic points. Since the map generating unit 120 registers the characteristic points on the map in order to detect the movement of the vehicle 1 on the parking route, it is advisable to focus on registering the characteristic points that are advantageous in detecting the movement. For example, when the vehicle 1 moves or turns on the route, a characteristic point that moves a lot on the camera image has a higher sensitivity and is more advantageous in detecting the movement of the vehicle 1 than a characteristic point that moves a small amount.

[0150] Since the map generating means collects information on characteristic points around the vehicle during learning driving, when registering another route on the map, it is advisable to register at least information on characteristic points located near another parking position on the map. For example, when the parking position during learning driving is different from the parking position during automatic parking, the position of the characteristic points near the parking position changes more significantly on the camera image than that of the characteristic points far from the parking position, so that the parking position can be more accurately adjusted by focusing on registering the characteristic points near the parking position.

[0151] Registering feature points in a focused manner near a specific position may be referred to as increasing the allocation of points. For example, the map generating unit 120 may increase the allocation of points in an area in the normal direction of the end point of a straight section or in the tangential direction of the end point of a turning section, and may register feature points in a focused manner. If a parking position is added, the area to which points should be allocated increases, so the map generating unit 120 may allocate different points depending on whether a parking position is added or not.

[0152] Fig. 16 to Fig. 18 are diagrams showing an example of a registration process of feature points. In Fig. 16, the area Ax (x=1 to 6) shown by an oval is an area in which the points are increased on the map in which the route RA in Fig. 12 is registered. Ax in Fig. 18 is an area in which the points are increased on the map in which only the route RB in Fig. 13 is registered. Ax in Fig. 17 is an area in which the points are increased on the map in which the route RA and route RB in Fig. 13 are registered.

[0153] The necessity for additionally registering feature points will be explained. For example, consider a case where route RB is additionally registered on a map in which feature points of route RA in FIG. 12 and area Ax in FIG. 16 are registered, but no additional feature points are registered. In that case, for example, when turning around at end point 29, there is a risk that the left-right position and the attitude (direction) of the vehicle body will be inaccurate because there is no feature point in the traveling direction. Therefore, it is good to allocate (distribute) many feature points also in area A7, which is an area existing in the tangent direction at end point 29.

[0154] The feature points at the rear of the parking position 32 are also important for accurately controlling the left-right position and the attitude of the vehicle body. Therefore, it is recommended that the map generating unit 120 also place many feature points in the area A8 that exists in the travel direction of the section 31-32. In this way, if the map generating unit 120 places additional feature points in response to the addition of a parking route, the distribution of the area where many feature points are placed will be like Ax in Fig. 17, and the vehicle 1 will be able to automatically park with high accuracy regardless of whether the parking position 27 or 32 is selected.

[0155] In addition, when the total number of feature points to be registered in one map is fixed, increasing the number of parking positions reduces the density of the feature points, which may result in poor parking accuracy. Therefore, the map generating unit 120 may generate a different map for each parking position and register different feature points for each map. For example, the feature points shown in Ax in FIG. 16 are registered in a map in which the route RA is registered, and the feature points shown in Ax in FIG. 18 are registered in another map in which the route RB is registered. In this way, the decrease in the density of the feature points is suppressed, so that the parking accuracy can be maintained. In other words, during automatic parking, the position estimating unit 150 can maintain the accuracy of self-position estimation.

[0156] However, dividing a map into two does not necessarily double the amount of data. For example, consider the case where a map M1 including the feature point Ax and the route RA in FIG. 16 and a map M2 including the feature point Ax and the route RB in FIG. 18 are generated separately. In this case, the map generation unit 120 sets a link that jumps from the end point 24 of the map M1 to the end point 24 of the map M2, and a link that jumps from the end point 24 of the map M2 to the end point 24 of the map M1. This allows the state management unit 110 to select a parking route at the end point 24 (branch point).

[0157] The data of route RA and the data of route RB are partially the same as shown in FIG. 14, so they may be compressed as shown in FIG. 15 to reduce the amount of data. However, the amount of data for the routes is not large to begin with, so there is not much difference even if they are compressed. On the other hand, many of the feature points registered on map M2 overlap with the feature points registered on map M1. Since there are many feature points and the amount of data is large, if there are many overlapping feature points, the efficiency of using the storage capacity decreases. Therefore, the map generation unit 120 may avoid overlapping by storing the feature points present in areas A1, A2, A4, and A5 that are registered in the two maps in the shared data.

[0158] When reading map M1, the state management unit 110 may read out feature point data dedicated to map M1 (feature points existing in areas A3 and A6) and shared data (feature points existing in areas A1, A2, A4, and A5). When reading map M2, the state management unit 110 may read out feature point data dedicated to map M2 (feature points existing in areas A3 and A6) and shared data (feature points existing in areas A1, A2, A4, and A5).

[0159] As mentioned above, when there is shared data and dedicated data, the shared data and dedicated data may be read into RAM and merged, or the shared data and dedicated data may be placed in different areas of ROM and accessed without distinction.

[0160] Furthermore, when the registration process is completed, the state management unit 110 may or may not notify the occupant that the registration process is completed. This is because there is no need to request the occupant for approval or operation when the registration process is completed.

[0161] For example, when an IG-OFF operation is performed during the execution of the registration process by the map generation unit 120, the state management unit 110 may request a power supply circuit (not shown) to maintain the power supply to the parking assistance ECU. In this case, the state management unit 110 may turn off the power supply without notifying the driver when the registration process by the map generation unit 120 is completed. This is because the HMI device 20 and the like are no longer functioning due to the IG-OFF operation, and thus no notification is possible. In addition, if the IG-ON is in effect when the registration process is completed, the state management unit 110 may notify the driver by displaying, on the HMI device 20 and the like, "Map registration has been completed" or "Automatic parking can be performed using the learned route." In addition, if the registration process is completed but no notification is possible, the same notification may be made the next time the IG-ON is turned on.

[0162] So far, an example has been described in which, in the proposal process, another parking position candidate is displayed on the HMI device 20 or the like, and the parking position is set by the occupant's operation and approval. This method of manually setting an additional parking position is convenient because it allows the occupant to add a parking position without actually parking the vehicle. However, actually parking the vehicle is expected to set a parking position that matches the occupant's preferences and has an optimal distance from surrounding objects, and actually parking the vehicle is also likely to satisfy the occupant. Therefore, the state management unit 110 may set a parking position by having the occupant park the vehicle manually.

[0163] (Continued learning run) From here, the continuous learning drive will be explained. The continuous learning drive is a learning drive that starts after parking in the learning drive. In other words, the continuous learning drive starts from the parking position where the vehicle 1 was parked in the learning drive. The map generating means sets the parking position when the occupant of the vehicle 1 manually drives the vehicle 1 out of the garage and parks the vehicle in a position different from the parking position from which the vehicle was parked as another parking position. Then, another route is generated from the parking start position of the learning drive to the other parking position, and the generated other route is registered in the map. For example, if there are spaces available for parking multiple vehicles in the garage, multiple parking positions will be registered, but with the continuous learning drive, there is no need to return to the parking start position to start the learning drive, so multiple parking positions can be registered in a short time.

[0164] 19 and 20 are diagrams for explaining an example of continuous learning driving. For example, as shown in Fig. 19, after the vehicle 1 is parked at the first parking position 27 from the end point 23 via 24, 25, and 26, the state management unit 110 inquires the occupant by displaying a message such as "Do you want to register another parking position?" on the HMI device 20 or the like.

[0165] If the driver answers Yes, the state management unit 110 requests the driver to park manually by displaying, for example, "Please move the vehicle and park it in the next parking position" on the HMI device 20. In response to this, the driver parks the vehicle in the parking position 32 from the parking position 27 via the turning point 33 and the turning end point 34, as shown in Fig. 20. The state management unit 110 then identifies the second parking position 32 based on the driver's manual parking.

[0166] In the above case, the map generation process has progressed from P001: start point processing → P002: collection processing → P003: end point processing → P004: analysis processing → P005: proposal processing before the start of continuous learning driving, and a request for continuous learning driving is made during the proposal processing. Then, after the start of continuous learning driving, the map generation process progresses as follows: P002: collection processing → P003: end point processing → P004: analysis processing → P005: proposal processing → P006: registration processing. The content of each process is roughly the same as the above-mentioned case of manual setting without actually parking, so mainly the differences will be explained.

[0167] In the continuous learning drive, a collection process is first performed in order to collect feature point information on the route to the second parking position 32 and to register feature points that are more advantageous for detection on the map.

[0168] The second collection process may be the same as the collection process on the route to the first parking position 27. The map generation unit 120 handles the feature point information collected in the second collection process without distinguishing between the feature point information collected in the first collection process. The state management unit 110 performs the second end point process when the gear position becomes P, and the notification unit 180 notifies the occupant to wait. The map generation unit 120 starts the second analysis process at the same time as the notification by the notification unit 180.

[0169] Before the second analysis process, the map generating unit 120 receives from the driving control unit 160 a movement route between the first parking position 27 and the second parking position 32 (a route from the first parking position 27 to the second parking position 32 via the end points 33 and 34). This route is a movement route illustrated by broken lines. Here, the map generating unit 120 specifies the relative positional relationship between the first parking position 27 and the second parking position 32 based on the movement route. However, the map generating unit 120 does not perform a process of approximating the received movement route (illustrated by broken lines) with straight sections and turning sections. This is because the route from the first parking position 27 to the second parking position 32 is an unnecessary route that is not used when performing automatic parking.

[0170] In the second analysis process, the map generating unit 120 calculates a parking route RB as shown in FIG. 20 based on the relative positional relationship between the first parking position 27 and the second parking position 32 and the parking route RA for parking at the first parking position 27. As described above, the proposal process shows the occupant an overhead view, allows the occupant to set an additional parking position on the screen, and the analysis process calculates the parking route from the parking start position to the additional parking position. However, among these processes, the process of calculating the parking route from the parking start position to the additional parking position can be performed regardless of the method of setting the additional parking position. In other words, once the second parking position 32 is set by continuous learning driving, the subsequent route calculation may be the same as when it is set manually. Therefore, to avoid duplication, the description will be omitted.

[0171] The next suggestion process may be the same as the first suggestion process performed before the start of the continuous learning drive, so a description of this will also be omitted. The final registration process is also generally the same as in the case of manual setting.

[0172] When the vehicle is driven for continued learning, in the registration process, the map generating unit 120 registers the feature point information collected in the first and second collection processes on the map. This differs from the case of manual setting in which the collection process is performed only once. In other words, the map generating means collects information on feature points around the vehicle when the vehicle occupant drives the vehicle manually after the learning drive, and when registering another route on the map, registers at least information on feature points located near another parking position on the map.

[0173] According to the principle of triangulation, when the coordinates of a feature point are identified, the accuracy of the coordinates is higher when the feature point is identified closer to the feature point than when the feature point is identified farther away from the feature point. For example, the accuracy of the coordinates of the feature point in the vicinity of the second parking position 32 is higher when the coordinates are identified during the continuous learning run in which the vehicle is parked at the second parking position 32 than when the coordinates are identified during the first learning run. In addition, a feature point that was not detected in the first collection process may be detected in the second collection process, and the newly detected feature point may be advantageous in estimating the position of the vehicle on a new parking route. Therefore, it is expected that the accuracy of the parking position when automatically parking at the second parking position 32 will be higher when the feature points collected during the continuous learning run are registered than when the feature points collected during the first learning run are registered.

[0174] Adding a parking position increases the area in which many feature points should be placed, just like in the case of manual setting. The registration process for registering the newly collected feature points on the map can be the same as in the case of manual setting, so a description thereof will be omitted.

[0175] Although the above describes an example in which the continued learning drive is started immediately after the initial learning drive, the continued learning drive does not necessarily have to be started immediately after the initial learning drive. For example, the occupant may start the continued learning drive on another day.

[0176] Specifically, if the vehicle 1 has not moved after the first learning drive and the data collected in the first collection process has not been erased, the occupant can start continued learning drive even if the vehicle's IG is turned OFF / ON during the drive. For example, when the vehicle 1 is parked during the first learning drive, the analysis process begins, but if the vehicle's IG is turned OFF before the analysis process is completed, the state management unit 110, together with the notification unit 180, performs a suggestion process the next time the vehicle 1's IG is turned ON. At this point, the occupant can decide whether to continue learning drive.

[0177] Furthermore, the map generating unit 120 may execute the registration process for registering the map during the IG-OFF period, assuming that the registration of the map data by the first learning drive is approved. When continuing learning drive is performed after the map is registered, the map generating unit 120 can generate a map by adding the feature points collected during the continued learning drive or the route data based on the route and feature points registered in the map, even if the data on the RAM collected in the first drive is lost, and register the map by overwriting the previously registered map. Alternatively, a map in which the feature points collected during the continued learning drive and the branch road starting from the end point 24 are registered may be generated separately from the previously registered map, and additionally registered. The additionally registered map may quote the feature points and routes of the previously registered map.

[0178] If the map from the first learning drive has been registered at the time of the second route calculation, the reference route referred to in the route calculation is the route registered in the map. Conversely, if the map has not been registered, the reference route is a set route that has not been registered in the map. Therefore, when generating another route to another parking position, the map generating means may be configured to generate the other route based on the route registered in the map or the set route, and may be configured to allow the reference route to be selected. In this way, using an existing route can reduce the time required for route calculation.

[0179] (Branch learning run) Next, the branch learning driving will be described. The branch learning driving is a learning driving that starts in the middle of automatic parking. The map generating means sets the parking position when the occupant of the vehicle 1 switches to manual driving in the middle of automatic parking and parks in a position different from the target parking position of automatic parking as another parking position. Then, another route from the start position of automatic parking to the other parking position is generated, and the generated other route is registered in the map. According to this branch learning driving, when it is not possible to park in the target parking position of automatic parking, it is possible to start the learning driving on the spot, and since there is no need to return to the parking start position to start the learning driving, it is possible to add a parking position in a short time.

[0180] Here, an example of branching learning driving will be described, in which the vehicle switches to manual driving midway through automatic parking while driving along the parking route shown in Fig. 12. This example is based on the premise that a complete learning driving has been performed in the past, in which the learning driving starts at end point 23, the vehicle passes through end points 24, 25, and 26, the vehicle is parked at parking position 27, and a map including parking route RA is registered. If the driver starts automatic parking at end point 23, and while the vehicle is automatically driving according to the map, the driver applies the brakes to stop the vehicle, switches to manual driving, and parks, this becomes branching learning driving.

[0181] For example, when the vehicle 1 stops during automatic driving in the section 23-24 of the parking route RA and starts manual driving, the map generating unit 120 may determine the position where the vehicle stops on the section 23-24 as a branch point. For example, when the occupant stops the vehicle 1 at the end point 24, the end point 24 becomes the branch point. Alternatively, the map generating unit 120 may compare the manual parking route with the automatic parking route, identify the position where the manual parking route branches off from the automatic parking route, and determine the branch point. For example, when the vehicle 1 stops on the section 23-24, stops automatic parking, starts manual driving, and then travels straight beyond the end point 24 as in the route RB of FIG. 13, the map generating unit 120 may determine the end point 24, which is the point where the route branches off, as a branch point. In the following description, the branch point is assumed to be the end point 24. For example, the map generating unit 120 additionally registers a branch road starting from the branch point 24 in the map.

[0182] In the above-described continuous learning driving, the vehicle 1 leaves the first parking position and parks in the second parking position, and does not drive on a route that goes directly from the parking start position to the second parking position. In addition, in the manual setting, the vehicle 1 does not park in the second parking position in the first place. Therefore, in the continuous learning driving and manual setting, the route for parking in the second parking position is generated based on the route for parking in the first parking position. In contrast, in the branch learning driving, the vehicle 1 drives on a route that goes directly from the parking start position to the second parking position, so there is no need to generate an additional route based on an existing route. In other words, the map generating unit 120 may register the route manually driven from the branch point 24 as a branch road in the map as it is. Alternatively, the route calculation process may be performed to approximate the learning driving route and set a route that generally follows the route, and a branch road consisting of a small number of sections may be registered in the map.

[0183] However, in the case of branch learning driving, as in the case of continuous learning driving and manual setting, a route for parking at the second parking position may be generated based on a route for parking at the first parking position and registered in the map. In other words, there is no need to change the route calculation method depending on the setting method of the second parking position, and the route for parking at the second parking position may always be generated based on a route for parking at the first parking position. In addition, in the case of branch learning driving, the route for parking at the first parking position is a route registered in the map, but in the case of continuous learning driving, the route for parking at the first parking position has been set but has not been registered in the map. Therefore, when generating another route, the map generating means generates another route based on the reference route, and the reference route can be selected from the route registered in the map or the route that has already been set. In this way, by using an existing route, the time required for route calculation can be shortened.

[0184] When the occupant of the vehicle 1 switches from automatic parking to manual driving, the notification regarding registering another route in the map is suppressed or not issued, compared to when the occupant of the vehicle starts manual driving from the parking position. The reason for this is that when automatic parking is switched to manual driving, the purpose is often to avoid oncoming vehicles or obstacles, and it is rare for the purpose to park in a different position. For this reason, it is better for the state management unit 110 to suppress or not issue the notification so as not to bother the occupant as much as possible.

[0185] For example, when the occupant stops the vehicle 1, stops automatic parking, and starts manual driving, the driver should not inquire about adding a parking route or setting a parking position. If the driver inquires, an unnecessary notification will be made when the purpose is to avoid an accident, and the notification will bother the driver. Therefore, the state management unit 110 causes the notification unit 180 to output an image displaying "A route for manual parking can be registered" and the like, and starts a collection process so that the driver can additionally register a parking position when he or she wishes. The notification when manual driving starts should be limited to displaying a text message on the screen of the HMI device 20, and not be notified by voice. In other words, when the parking assistance device interrupts automatic parking and starts manual driving, it starts the learning driving process silently.

[0186] In the above case, the status management unit 110 causes the map generation unit 120 to start the learning drive processing without asking the occupant whether to add a parking position, so it can be said that the learning drive processing is started speculatively.

[0187] Furthermore, when the map generating unit 120 detects that the gear position has become P in the end point process, it may start the analysis process and subsequent processes without notifying the occupant, and may execute the registration process in the background without seeking the occupant's approval. In other words, all processes including the map registration may be executed speculatively. In such a case, the state managing unit 110 may notify the occupant after the registration process is completed or when the ignition is turned on the next time that the parking position has been added and that the last registered parking position can be deleted.

[0188] As described above, when switching to manual parking during automatic parking, it is more likely to be for other purposes such as avoiding obstacles, loading and unloading luggage, or for passengers to disembark, rather than for the purpose of adding a parking position. For this reason, the state management unit 110 may issue a notification in a manner that allows the vehicle to be left unattended, rather than in a manner that does not allow the vehicle to proceed unless approval or instructions are given.

[0189] For example, the state management unit 110 may display a text message such as "Do you want to register the last manually parked position?" on the HMI device 20, etc. If "Yes" is not entered within 5 seconds, the map generation unit 120 may delete the map of the branch road. In this way, the driver does not have to do anything if he does not want to add a parking position.

[0190] The map generating unit 120 may also mark the map as "unapproved" and retain the registration. In this case, the state managing unit 110 may display a list of unapproved maps and request approval the next time manual parking is performed. The map generating unit 120 may also automatically delete the unapproved map with the oldest registration date when the number of registered maps increases.

[0191] (Additional learning run) Next, the additional learning run will be described. The additional learning run is a learning run that starts after automatic parking. The map generating means sets the parking position when the vehicle 1 leaves the parking position where the vehicle 1 was parked by automatic parking by manual driving and drives, and parks at a position different from the parking position of the automatic parking, as another parking position. Then, another route from the parking start position of the automatic parking to the other parking position is generated, and the generated other route is registered in the map. When there are spaces available for parking a plurality of vehicles in the garage, one parking position may be registered, and after automatic parking is tried, the next parking position may be registered. In that case, according to the additional learning run, it is not necessary to return to the parking start position to start the learning run, so that the next parking position can be registered in a short time. The additional learning run may be regarded as a continuous learning run that starts at the parking position of the automatic parking, not at the parking position of the learning run, and a branch learning run that starts at the parking position of the automatic parking, not during the automatic parking.

[0192] The notification of the additional learning drive may be the same as the notification of the continuous learning drive. For example, in the example of Fig. 12, when the vehicle 1 is automatically parked in the parking position 27, the state management unit 110 displays a message on the HMI device 20 etc. asking "Do you want to register another parking position?" In this case, if the occupant answers Yes, the state management unit 110 displays a message on the HMI device 20 etc. asking the occupant to park manually by displaying "Please move the vehicle and park it in the next parking position" or the like.

[0193] In response to this, the driver manually drives the vehicle out of the parking lot and parks the vehicle in parking position 32, as in the example of FIG. 19. In this case, the map generation unit 120 identifies the second parking position 32 based on the manual parking by the driver. In the case of additional learning driving, the method of setting branch points and the process of generating routes for branching roads may be generally similar to those in continuous learning driving, but the data used as the basis for route generation is different. The data used as the basis for route generation in the case of additional learning driving is the automatic parking data, as in the case of branching learning driving.

[0194] In route generation after the additional learning drive is completed, a route to park at the second parking position is generated based on the positional relationship between the first parking position and the second parking position and the route to park at the first parking position. In the case of continued learning drive, the route data in memory is the route data before registration. In the case of additional learning drive, the route data in memory is the route registered on the map. However, since the content of the data is the same, there is no need to distinguish between the two.

[0195] In addition, if there is an error in the vehicle control during automatic parking and there is a difference between the target parking position (end point of the route) registered in the map and the actual parking position, this may be compensated for during route generation. For example, the map generation unit 120 first calculates the positional relationship between the first parking position and the second parking position from the route of the additional learning drive. By adding the data of the difference between the actual parking position and the target parking position to the data of this positional relationship, the accumulation of errors is suppressed. Therefore, the storage unit 170 may store data of the difference between the actual parking position and the target parking position when the vehicle is automatically parked in preparation for the additional learning drive. In this way, the map generation unit 120 can suppress the accumulation of errors even when the additional learning drive is performed on a day other than the day of automatic parking. Other explanations of the process of route calculation based on the second parking position will be omitted because they overlap with the explanation for the case of continuous learning drive.

[0196] The processing of the feature points of the additional learning drive may be the same as the processing of the feature points of the continuous learning drive. In addition, as an application in the case of adding a parking position by a learning drive, the map generating unit 120 may exclude the feature points registered in the existing map from the target of the collection process. Alternatively, the map generating unit 120 may exclude, during the registration process, the feature points registered in the existing map from the target of the processing among the feature points collected in the collection process. By reducing the target of the processing in this way, the processing time can be shortened.

[0197] We have explained above how to separate maps for parking route RA and parking route RB, and how to combine maps into one, but in the case of additional learning travel, it is better to additionally register a map of the branch roads in addition to the map of the parking route RA. In the case of additional learning travel, the branch roads are often not registered, so if the map of the branch roads is separated, it becomes easier to delete them.

[0198] For example, as shown in Fig. 16, if there is already a map on which route RA for parking at parking position 27 and the feature points of areas A1 to A6 are registered, and then as shown in Fig. 18, a map on which route RB for parking at parking position 32 and the feature points of areas A1, A2, A4, A5, A7, and A8 are registered is added, the two maps can be treated independently. Although it is necessary to change the links of the existing map to accommodate branching, there is no need to change the feature points, so the amount of change to the map can be kept to a minimum.

[0199] If the map is not separated, the feature points registered on the map will spread over the range of areas A1 to A8 in Fig. 17, resulting in more feature points than when a single parking route is registered. In the example of Fig. 17, the car is parked in the same garage, but in the case of additional learning drives, the car is often temporarily parked in a completely different location outside the garage and an additional parking location is not registered, so the additional registered feature points are often wasted.

[0200] If the map is not separated and the amount of map data is kept constant, the map generating unit 120 needs to select and delete some of the feature points already registered on the map in order to register the feature points collected in the second learning drive on the map. This has the disadvantages of increasing the amount of processing and lengthening the processing time, and deleting the feature points registered on the map, degrading the parking accuracy. Furthermore, if an additional parking position is not registered, the added feature points become unnecessary, but the deleted feature points cannot be restored, so the accuracy continues to deteriorate. If the map generating unit 120 generates a map of the branching roads separately, there are no disadvantages such as increased processing time and deterioration of accuracy in exchange for an increase in the amount of data.

[0201] So far, a method for generating a map including a plurality of parking positions has been described. When performing automatic parking using the map, a parking position is selected by some means. For example, the occupant may select a parking position at the start of automatic parking or during automatic parking, but it is preferable to assign priorities to the parking positions in advance. In this way, the parking assistance ECU 100 can automatically select a parking position based on the priorities, so that the occupant does not need to give instructions during automatic parking. Any means may be used to determine the priorities. For example, the order in which the parking positions are registered may be used as the priorities, or the occupant may manually determine the priorities. In the following example of automatic parking, priorities are assigned in order of proximity to the parking start position.

[0202] (Automatic parking processing) From here, an example of the process executed by the parking assist ECU 100 during automatic parking will be described. Fig. 21 is a flowchart showing an example of the process executed by the parking assist ECU 100 according to the embodiment during automatic parking. Fig. 22 is a diagram explaining an example of the automatic parking executed by the parking assist ECU 100 according to the embodiment.

[0203] Hereinafter, an example of automatic parking using a map including a plurality of parking positions will be described as an example of processing executed by the parking assistance ECU 100 according to the embodiment. The map generated by the learning drive stores a route as shown in FIG.

[0204] The map generated during the learning drive stores route R-1 for parking at parking position PP-1, route R-2 that branches off from route R-1 at branch point B-1 for parking at parking position PP-2, and route R-3 that branches off from route R-2 at branch point B-2 for parking at parking position PP-3.

[0205] If the number of stored routes is M, in the example of FIG. 22, M=3. In the example of FIG. 22, the automatic parking system travels along route RN (N=1-3) with parking position PP-N (N=1-3) as the target parking position. Then, when the vehicle reaches branch point BN (N=1-2), if the vehicle cannot be parked at parking position PP-N (N=1-2), the system changes the route to N=N+1. However, if the vehicle cannot be parked at parking position PP-M when N=M (M=3), the system automatically stops and ends the automatic parking.

[0206] 21 is managed by the state management unit 110. The parameters (M, N) used in the processing are also managed by the state management unit 110. The state management unit 110 also determines the progress of each step and jumps from step to step.

[0207] First, the state management unit 110 reads a map from the storage unit 170 (step S001). The storage unit 170 can store multiple maps, and the map to be read is selected by the state management unit 110. For example, the state management unit 110 may obtain the position information of the vehicle from the navigation device 40 via the in-vehicle LAN, compare the start position coordinates of each map, and select the map with the closest start position.

[0208] Moreover, the state management unit 110 sets the initial values ​​of the parameters based on the information recorded in the map read from the storage unit 170 (the number of parking positions=3) (M=3, N=1).

[0209] The space recognition unit 140 constantly detects whether there is an obstacle in the traveling direction of the vehicle 1 (step S002). For example, the space recognition unit 140 detects the front based on an image from the front camera 2c. The state management unit 110 determines whether there is an obstacle within a predetermined range in the traveling direction of the vehicle 1 based on the detection result. The obstacle detection means may be a sonar or a short-range radar (not shown).

[0210] If an obstacle is detected (step S002: Yes), the state management unit 110 instructs the driving control unit 160 to automatically stop the vehicle 1 (step S003). In other words, if the obstacle is detected at the start of automatic parking, the vehicle 1 does not start, and if the vehicle is traveling, the vehicle 1 automatically stops. If an obstacle is not detected (step S002: No), the process proceeds to step S004.

[0211] The state management unit 110 instructs the driving control unit 160 to drive the vehicle 1 along the route RN (step S004). Even if the vehicle 1 has automatically stopped in step S003, if no obstacle is detected (step S002: No), the vehicle 1 resumes driving.

[0212] Next, the state management unit 110 determines whether the branch point BN has been reached (step S005). If the branch point BN has not been reached (step S005: No), the process returns to step S002 and the loop of steps S002-S005 is repeated.

[0213] If the branch point BN is reached (step S005: Yes), the state management unit 110 determines whether parking is possible at the parking position PP-N (step S006). For example, when the vehicle is at the branch point B-1, the state management unit 110 may determine that parking is possible if the space recognition unit 140 detects the left front of the vehicle 1 based on the image from the left side camera 2a and all detection points of the parking position PP-1 are detected among the characteristic points registered on the map.

[0214] In addition, if none of the multiple characteristic points located behind parking position PP-N among the characteristic points registered on the map are detected, the status management unit 110 may determine that parking is not possible at parking position PP-N.

[0215] The state management unit 110 may also detect the direction of parking position PP-N using a side sonar (not shown), and if the distance to the nearest detected obstacle corresponds to the distance to the back wall of parking position PP-N, determine that parking is possible at parking position PP-N. Alternatively, the state management unit 110 may temporarily stop the vehicle 1 at branch point BN, and inquire of the occupant as to whether parking is possible at parking position PP-N, and have the occupant input the result.

[0216] If parking is not possible (step S006: No), the state management unit 110 determines whether the route being traveled is the last route registered in the map (step S007). In the example of FIG. 21, the number of routes M is 3, so if it is not the last route (N=3), the process proceeds to step S009. If the last route R-3 being traveled is being traveled, the process proceeds to step S008.

[0217] Step S008 is a process to be performed when the vehicle cannot be parked at any of the parking positions registered on the map. The state management unit 110 instructs the driving control unit 160 to stop the vehicle 1 (step S008), and ends this process.

[0218] For example, a notification process may be added to step 008. In this case, the state management unit 110 may notify the occupant by displaying a message such as "Automatic parking will end because the vehicle cannot be parked in any of the registered parking positions" on the HMI device 20, etc. Also, the state management unit 110 may further add a step of confirming that the gear position has become P, thereby ensuring that the vehicle 1 does not move after automatic parking ends.

[0219] If it is not the last route in step S007 (step S007: No), the state management unit 110 sets N=N+1 and changes the route to be traveled from route RN to route R-N+1 (step S009). Step S009 is a process for continuing automatic parking by changing the route.

[0220] For example, if the route being traveled is route R-1 (N=1), the state management unit 110 branches from route R-1 to route R-2, and if the route being traveled is route R-2 (N=2), the state management unit 110 branches from route R-2 to route R-3. After branching, the process returns to step S002. Steps S002 to S009 are a process loop for repeating branching, and in this map, branching can occur up to two times.

[0221] If parking is possible in step S006 (step S006: Yes), the state management unit 110 determines whether or not there is an obstacle within a predetermined range in the traveling direction of the vehicle (step S010).

[0222] If there is an obstacle in the traveling direction (S010: Yes), the state management unit 110 instructs the traveling control unit 160 to automatically stop the vehicle 1 (step S011). If there is no obstacle in the traveling direction (step S010: No), the state management unit 110 instructs the traveling control unit 160 to resume traveling on the route RN (step S012).

[0223] In addition, the state management unit 110 determines whether the vehicle 1 has reached the parking position PP-N (step S013). If the vehicle 1 has not reached the parking position PP-N (step S013: No), the state management unit 110 continues traveling until the vehicle 1 reaches the parking position PP-N.

[0224] If the vehicle 1 reaches the parking position PP-N in step S013 (step S013: Yes), the state management unit 110 automatically stops the vehicle 1 (step S014) and ends the automatic parking control. Note that the state management unit 110 may further add a step of activating the parking brake before ending the control, thereby ensuring that the vehicle 1 does not move after the automatic parking ends.

[0225] 21 includes a process for automatically stopping the vehicle 1 when an obstacle is detected. However, the state management unit 110 may also process the vehicle 1 when the occupant applies the brakes in the same manner as when an obstacle is detected. The state management unit 110 may also resume automatic parking of the vehicle 1 when the occupant releases the brakes.

[0226] In FIG. 21, an example is described in which the state management unit 110 instructs the driving control unit 160 to stop the vehicle 1 and end the automatic parking when the vehicle cannot be parked at any of the parking positions registered in the map. In this case, the state management unit 110 may display a message such as "Please park manually because the vehicle cannot be parked at any of the registered parking positions" on the HMI device 20 or the like via the notification unit 180 to prompt the occupant to park manually. In addition, the notification means (the state management unit 110 and the notification unit 180) may further notify the occupant that the parking position where the vehicle has been manually parked is to be registered on the map. This notification means includes the state management unit 110 and the notification unit 180, and may also include the HMI device 20. If the occupant performs manual parking in response to this notification, this becomes branch learning driving.

[0227] The notification regarding the additional registration of the parking position may be made not only when the vehicle cannot park at all parking positions. In the case of FIG. 22, the notification may be made before the vehicle becomes stuck beyond the branch point B-2. For example, when the vehicle occupant stops the vehicle or when the vehicle cannot be parked at the parking position registered in the map, the notification means suggests to the vehicle occupant that automatic parking should be switched to manual driving to park. In this way, the occupant can start branch learning driving at any time and from any position without waiting for automatic parking to become stuck. When making this suggestion, the occupant may be notified that a parking position can be additionally registered by manual driving. In this way, even an occupant who is unfamiliar with the parking assistance function can additionally register a parking position.

[0228] For example, if there are parking positions for multiple vehicles in a garage, it is considered that the occupant will reach the same decision before the parking assistance device determines that it is not possible to park in all of the parking positions and automatically stops the vehicle. Therefore, for example, from the point in time when it is determined that it is not possible to park in the first target parking position, a button for selecting to switch to manual driving and park the vehicle is displayed. In addition, since the occupant may determine that it is not possible to park the vehicle before the space recognition unit 140 starts detecting the parking position, the button may start to be displayed when the occupant stops the vehicle. In this way, the occupant can immediately start manual driving, so that the parking position can be smoothly registered as an additional parking position.

[0229] The means for instructing the start of manual parking is not limited to a button, but may be the operation of the brake, steering wheel, turn signal, or hazard lamp. In addition, it is not necessarily necessary to stop the vehicle when starting manual parking. For example, if the occupant operates the steering wheel during automatic parking to deviate from the parking route and park in a different location, the driving from the position where the steering wheel operation was started to the parking location may be regarded as a branch learning driving, and the parking route may be additionally registered, so that the additionally registered parking route can be used the next time automatic parking is performed.

[0230] From here, the control of automatic parking will be explained, focusing on the processing of data. For example, the data processing of automatic parking may be expressed as follows. As shown in FIG. 14 above, the route data stored in the map is a collection of section data. The state management unit 110 reads out the section data, and instructs the driving control unit 160 to drive based on the steering angle and travel distance values ​​recorded therein. The state management unit 110 repeats this process for each section, thereby driving along the recorded parking route.

[0231] For example, the state management unit 110 reads data for a section using a pointer. The data for a section has a link that indicates the start address of the data for the next section. The state management unit 110 can read data for a series of parking routes by following this link and reading the data for the next section.

[0232] In such a configuration, the branching process corresponds to the process of changing the link to be followed. For example, assume that the data for a section has a link indicating the start address of the data for the next section, and a link indicating the start address of the data for the first section of the branching road. When the state management unit 110 follows the latter link and reads out the data for the section, it can obtain the steering angle when traveling on the branching road and the travel distance for that section.

[0233] The state management unit 110 then follows the links to the next sections one by one, and when it reaches link "0" indicating the end point, that is the final section. The state management unit 110 stops the vehicle when the final section is completed, and ends the automatic parking.

[0234] Here, the state management unit 110 selects whether or not to branch at a branch point depending on whether parking is possible at a parking location connected to the route. Therefore, it may be said that the state management unit 110 functions as a route selection means that selects a route at a branch point. In addition, since the determination of whether parking is possible is performed by the space recognition unit 140, the space recognition unit 140 may be included in the route selection means.

[0235] In summary, the embodiment of the present disclosure is a parking assistance device that includes a map generating means for registering a map including a route from a parking start position to a parking position during learning driving, and a driving control means for autonomously driving a vehicle based on the map during automatic parking, the map having multiple routes from one parking start position to multiple parking positions that branch off along the way, and a route selecting means for selecting a route for the vehicle to travel from the multiple routes during automatic parking. By using such a parking assistance device, even if the planned parking position at the start of automatic parking is blocked, the route selecting means can change the route along the way and park the vehicle in another parking position.

[0236] The route selection means may detect a parking position or a parking route when the vehicle approaches a branch point, and select a route according to the detection result. In this case, the vehicle automatically selects an available parking position and automatically parks, so the driver does not need to instruct the automatic parking. For example, the spatial recognition unit 140 analyzes the motion parallax of the image of the parking space when the vehicle approaches a parking position. If there is a parallax larger than when the floor of the parking space is captured, the spatial recognition unit 140 may determine that parking is not possible because the floor is not visible, and may branch off to a branch road leading to another parking position.

[0237] For example, the route selection means may select a route according to the detection result of the image processing unit 130. Specifically, the image processing unit 130 detects characteristic points at the back of the parking space or on the floor of the parking space, which are registered on a map. If the image processing unit 130 does not detect the corresponding characteristic points, it may determine that the parking space is not visible and therefore parking is not possible, and may branch to a branch road leading to another parking position.

[0238] For example, the route selection means may select a route according to the detection result of an obstacle detection device such as a sonar or radar. Specifically, the obstacle detection device detects whether there is an obstacle in the direction of the parking position. If the distance to the detected obstacle is shorter than the distance to the parking position, the obstacle detection device may determine that parking is impossible due to the presence of an obstacle, and may branch to a branch road leading to another parking position.

[0239] The route selection means may select a route according to an instruction from the occupant. This may be rephrased as the route selection means inquiring of the occupant of the vehicle regarding a parking position when the vehicle approaches the branch point, and selecting a route for the vehicle to travel according to the occupant's response. In other words, a user interface that allows the occupant to select a parking route may be the route selection means. In this way, it becomes possible to change the parking position according to the occupant's preferences and convenience. Furthermore, if the occupant selects the parking position, it becomes unnecessary to detect the parking position, and therefore the price of the parking assistance device can be reduced.

[0240] For example, the state management unit 110 may stop the vehicle 1 at a branch point and output a voice message to the HMI device 20, such as "Do you want to park in the parking spot on the left? Or go straight?" In this case, the state management unit 110 may select a route by receiving a button displayed on a touch panel from the occupant. Alternatively, a button for selecting a branch may be displayed while the vehicle is traveling, and if the button is pressed before the branch point, the route will branch, and if the button is not pressed, the route will not branch.

[0241] The state management unit 110 may also notify the occupant of the result of the parking route selected by the device such as the space recognition unit 140 via the notification unit 180. Furthermore, the state management unit 110 may allow the occupant to change the route selection when the notification result differs from the occupant's judgment.

[0242] For example, when the state management unit 110 issues a notification such as "Pass the parking position on the left and go straight ahead," the occupant may think, "There is no problem with the parking position on the left, and I would like to park in the parking position on the left." In this case, the occupant may instruct the route change by, for example, gently turning the steering wheel to the left. Then, when the state management unit 110 receives a steering wheel operation by the occupant, it may overturn the determination, select a route on the left, and simultaneously notify the occupant, "Park on the left." At that time, the state management unit 110 may notify the occupant that the instruction has been received by vibrating the steering wheel, for example. This can prevent the occupant from steering more than necessary.

[0243] Although the embodiment of the present invention has been described above, the above-mentioned embodiment is presented as an example and is not intended to limit the scope of the present invention. This new embodiment can be embodied in various other forms. In addition, various omissions, substitutions, and modifications can be made without departing from the gist of the invention. In addition, this embodiment is included in the scope and gist of the invention, and is included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0244] 1 vehicle 2 Camera 10 Operating device 20 HMI device 30 Vehicle control device 40 Navigation devices 50 Sonar ECU 100 Parking Assist ECU 101 CPU 102 ROM 103 RAM 104 I / O 105 IMP 106 Communication I / F 110 Status Management Unit 120 Map Generation Unit 130 Image processing section 140 Spatial recognition section 150 Position estimation part 160 Driving control unit 170 Storage section 180 Notification Department

Claims

1. A map generating means for registering a map including a route from a parking start position to a parking position during learning travel; A driving control means for autonomously driving the vehicle based on the map during automatic parking; Equipped with the map has a plurality of routes from a single parking start location to a plurality of parking locations; a route selection means for selecting a route along which the vehicle will travel from the plurality of routes during automatic parking; Further comprising: Parking assistance device.

2. The map generating means When a parking position different from the parking position is set after the learning travel, generating a different route from the parking start position of the learning drive to the different parking position, and registering the generated different route in the map; 2. The parking assistance device according to claim 1.

3. The map generating means A parking position where the vehicle is parked by a vehicle occupant manually driving the vehicle from a parking position where the vehicle is parked during the learning drive is set as the different parking position.

3. The parking assistance device according to claim 2.

4. The map generating means The vehicle occupant switches the automatic parking to manual driving, and sets the parking position when the vehicle is driven and parked by manual driving as the different parking position.

3. The parking assistance device according to claim 2.

5. The map generating means A parking position where the vehicle is parked by the automatic parking and then parked by an occupant of the vehicle through manual driving is set as the different parking position.

3. The parking assistance device according to claim 2.

6. The map generating means Collecting detection information around the vehicle during the learning drive; When the vehicle is parked during the learning drive, setting the alternative parking position based on the detection information; 3. The parking assistance device according to claim 2.

7. The device further includes a suggestion means, The suggestion means includes: When the vehicle is parked during the learning drive, Suggesting an additional parking position to an occupant of the vehicle, and setting the additional parking position based on an operation or approval of the occupant.

3. The parking assistance device according to claim 2.

8. The map generating means Collecting detection information around the vehicle during the learning drive, and setting parking position candidates based on the detection information, The suggestion means suggests the parking position candidate to an occupant of the vehicle.

8. A parking assistance device according to claim 7.

9. The map generating means When generating the alternative route, A route registered in the map or a route that has already been set is used as a reference route, Based on the route of the reference, generating said alternative route; 3. The parking assistance device according to claim 2.

10. The route data is a collection of data on sections constituting the route, The map generating means Based on the route of the reference, When generating the alternative route, generating data of the section of the other route by copying, processing, or sharing the data of the referenced section; 10. A parking aid according to claim 9.

11. the route data has a link for jumping to the data of the other route, The link is data of the other route stored in the same map as the map storing the route, or or data of the other route stored in a map other than the map storing the route.

11. A parking assistance device according to claim 10.

12. Further comprising a notification means, The notification means issues a notification regarding the registration of the alternative route on the map, When an occupant of the vehicle switches the automatic parking to manual driving, The notification is suppressed or not made, more so than when the occupant of the vehicle starts manual driving from the parking position.

3. The parking assistance device according to claim 2.

13. The map generating means During the learning drive, information on characteristic points around the vehicle is collected; When registering the alternative route on the map, registering information on at least characteristic points located near the different parking positions on the map; 3. The parking assistance device according to claim 2.

14. The map generating means collecting information on characteristic points around the vehicle when an occupant of the vehicle manually drives the vehicle after the learning drive; When registering the alternative route on the map, registering information on at least characteristic points located near the different parking positions on the map; 3. The parking assistance device according to claim 2.

15. the path includes a branch point; The route selection means When the vehicle approaches the branch point, an inquiry is made to an occupant of the vehicle regarding a parking position; Selecting a route for the vehicle to travel in response to the response of the occupant.

2. The parking assistance device according to claim 1.

16. the path includes a branch point; The route selection means Detecting a parking position or a parking path when the vehicle approaches the branch point; selecting a route along which the vehicle will travel in response to a result of the detection; 2. The parking assistance device according to claim 1.

17. Further comprising a notification means, The notification means includes: An occupant of the vehicle stops the vehicle, or If the vehicle cannot be parked in a parking position registered on the map, Propose to an occupant of the vehicle to switch the automatic parking to manual driving and park the vehicle.

2. The parking assistance device according to claim 1.

18. When an occupant of the vehicle performs a predetermined operation, the automatic parking is switched to manual driving, The predetermined operation is the operation of a predetermined button, or the operation of a brake, a steering wheel, a turn signal, or a hazard lamp.

5. A parking assistance device according to claim 4.

Images