Parking assistance method and parking assistance device

Abstract

In the parking assistance method of the present invention, the positions of multiple marker objects detected around the target parking position when the vehicle is parked at the target parking position are stored as learned marker object positions (S2), the number of times the learned marker object position is consistent with the position of the marker object detected around the vehicle when the vehicle is traveling near the target parking position after the learned marker object position is stored, that is, the surrounding marker object position, is counted for each of the multiple learned marker object positions, and a learned marker object position that is consistent with the surrounding marker object position more times is given a higher reliability than a learned marker object position that is consistent with the surrounding marker object position less times (S12, S25), and the learned marker object position above a specified reliability threshold among the learned marker object positions is compared with the positions of the marker objects detected around the vehicle, and the relative position of the vehicle relative to the target parking position is calculated (S22).

Description

Technical Field

[0001] The invention relates to a parking assistance method and a parking assistance device. Background Art

[0002] As a technology related to parking assistance for the vehicle to a target parking position, the following patent document 1 is known. In patent document 1, marker objects around the target parking position are detected and stored, the relative position of the vehicle to the target parking position is calculated based on the positions of the marker objects detected around the vehicle during automatic parking and the stored positions of the marker objects, and the vehicle is automatically moved to the target parking position based on the relative position.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Publication No. 2017-138664

[0006] Problems to be Solved by the Invention

[0007] However, in the case where the stored marker object is a movable object, the calculation accuracy of the host vehicle relative to the target parking position may be reduced due to a change in the position of the marker object after the marker object is stored. Summary of the invention

[0008] An object of the present invention is to suppress a decrease in the calculation accuracy of the relative position due to changes in the positions of stored marker objects in parking assistance for calculating the relative position of a vehicle with respect to a target parking position based on the positions of marker objects around the target parking position that are pre-stored.

[0009] In a parking assistance method of one embodiment of the present invention, the positions of multiple marker objects detected around the target parking position when parking the vehicle at the target parking position are stored as learned marker object positions, the number of times the learned marker object position is consistent with the position of the marker object detected around the vehicle when the vehicle is driving near the target parking position after the learned marker object position is stored, that is, the surrounding marker object position, is counted for each of the multiple learned marker object positions, a learned marker object position that is consistent with the surrounding marker object position more times is assigned a higher reliability than a learned marker object position that is consistent with the surrounding marker object position less times, a learned marker object position above a specified reliability threshold in the learned marker object position is compared with the positions of the marker objects detected around the vehicle, the relative position of the vehicle with respect to the target parking position is calculated, a target driving trajectory from the current position of the vehicle to the target parking position is calculated based on the calculated relative position, and parking assistance control is performed to assist the vehicle in moving along the calculated target driving trajectory.

[0010] Effects of the Invention

[0011] According to the present invention, in parking assistance for calculating the relative position of the vehicle with respect to the target parking position based on the positions of marker objects pre-stored around the target parking position, it is possible to suppress the reduction in calculation accuracy of the relative position due to changes in the stored marker object positions.

[0012] The objects and advantages of the present invention are embodied and realized by the elements and combinations thereof shown in the scope of the claims. The above general description and the following detailed description are only examples and illustrations and should not be understood as limiting the present invention as the scope of the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a diagram showing an example of a schematic configuration of a parking assist device.

[0014] Figure 2 It is a schematic diagram of the positions of landmark objects around the target parking position.

[0015] Figure 3 This is an explanatory diagram of an example of a parking assistance method.

[0016] Figure 4 (a) to (d) are explanatory diagrams of an example of a method of assigning reliability.

[0017] Figure 5 yes Figure 1 A block diagram of an example of the functional structure of a controller.

[0018] Figure 6This is a flowchart of an example of processing when storing the position of a marker object.

[0019] Figure 7 This is a flowchart of an example of processing when updating reliability.

[0020] Figure 8 This is a flowchart of an example of parking assist control according to the embodiment. DETAILED DESCRIPTION

[0021] (structure)

[0022] Figure 1 1 is a diagram showing an example of a schematic configuration of a parking assistance device according to an embodiment. A vehicle 1 includes a parking assistance device 10 for assisting the vehicle 1 in parking at a target parking position. The parking assistance device 10 calculates a target driving trajectory from the current position of the vehicle 1 to the target parking position, and assists the vehicle in driving along the target driving trajectory.

[0023] The parking assistance of the parking assistance device 10 includes various methods of assisting the vehicle 1 to travel along the target driving trajectory. For example, the parking of the vehicle 1 may be assisted by performing an automatic driving operation that controls the vehicle 1 so that the vehicle 1 travels along the target driving trajectory of the vehicle 1 to the target parking position. In addition, the automatic driving operation that controls the vehicle 1 so that the vehicle 1 travels along the target driving trajectory of the vehicle 1 to the target parking position means that the steering angle, driving force, and braking force of the vehicle are controlled in whole or in part to automatically implement the driving of the vehicle 1 along the target driving trajectory, thereby assisting the control of the parking operation of the occupant.

[0024] In addition, for example, the parking of the host vehicle 1 may be assisted by displaying the target travel trajectory and the current position of the host vehicle 1 on a display device that can be visually recognized by the occupant of the host vehicle 1 .

[0025] The parking assist device 10 includes a positioning device 11 , an object sensor 12 , a vehicle sensor 13 , a human-machine interface (HMI) 15 , an actuator 16 , and a controller 17 .

[0026] The positioning device 11 measures the current position of the host vehicle 1. The positioning device 11 may include, for example, a global positioning system (GNSS) receiver. The GNSS receiver may be, for example, a global positioning system (GPS) receiver.

[0027] The object sensor 12 detects objects within a specified distance range from the host vehicle 1 (e.g., the detection area of ​​the object sensor 12). The object sensor 12 detects the surrounding environment of the host vehicle 1, such as the relative position of the objects existing around the host vehicle 1 and the host vehicle 1, the distance between the host vehicle 1 and the objects, and the direction in which the objects exist. The object sensor 12 may include, for example, a camera that captures the surrounding environment of the host vehicle 1. The camera may also be, for example, a camera that captures the surroundings of the host vehicle 1 and generates a captured image that is transformed into a bird's-eye view (surround view monitor image). The object sensor 12 may include a ranging device such as a laser finder (LRF), a radar, a lidar (light detection and ranging), or a laser radar.

[0028] The vehicle sensor 13 detects various information (vehicle information) obtained from the host vehicle 1. The vehicle sensor 13 includes, for example, a vehicle speed sensor for detecting the running speed (vehicle speed) of the host vehicle 1, a wheel speed sensor for detecting the rotation speed of each tire provided on the host vehicle 1, a three-axis acceleration sensor for detecting acceleration (including deceleration) in three axial directions of the host vehicle 1, a steering angle sensor for detecting the steering angle of the steering wheel, a steering angle sensor for detecting the steering angle of the steering wheel, a gyro sensor for detecting the angular velocity generated in the host vehicle 1, and a yaw rate sensor for detecting the yaw rate.

[0029] The human-machine interface 15 is an interface device for transmitting and receiving information between the parking assisting device 10 and the occupant. The human-machine interface 15 includes a display device (for example, a display screen of a navigation system and a display device provided near the instrument in front of the driver's seat) that can be visually recognized by the occupant of the vehicle 1. In addition, the human-machine interface 15 includes an operating element for receiving an operation input of the occupant to the parking assisting device 10. For example, the operating element may be a button, a switch, a lever, a dial, a keyboard, a touch panel, etc.

[0030] The actuator 16 includes a steering actuator, an accelerator actuator, and a brake control actuator. The steering actuator controls the steering angle of the steering mechanism of the vehicle 1 according to the control signal from the controller 17. The accelerator actuator controls the accelerator opening of the engine or the drive motor, i.e., the drive device according to the control signal from the controller 17. The brake actuator activates the brake device according to the control signal from the controller 17.

[0031] The controller 17 is an electronic control unit that performs parking assist control of the host vehicle 1. The controller 17 includes peripheral components such as a processor 18 and a storage device 19. The processor 18 may be, for example, a CPU or an MPU.

[0032] The storage device 19 may include a semiconductor storage device, a magnetic storage device, an optical storage device, etc. The storage device 19 may include a register, a cache memory, a ROM and a RAM used as a main storage device, etc. For example, the processor 18 executes a computer program stored in the storage device 19 to realize the functions of the controller 17 described below.

[0033] In addition, the controller 17 may be formed by dedicated hardware for executing each information processing described below. For example, the controller 17 may also include a functional logic circuit set in a general-purpose semiconductor integrated circuit. For example, the controller 17 may include a programmable logic device such as a field programmable gate array.

[0034] Next, an example of parking assist control by the parking assist device 10 will be described. Figure 2 . When parking assistance using the parking assistance device 10 is started, first, the relative position relationship between the marker object position of the marker object existing near the target parking position 2 and the target parking position 2 is stored in the storage device 19. The target parking position 2 refers to the target position for parking the vehicle 1. The marker object refers to a ground object used to mark the current position of the vehicle 1. The marker object can be, for example, a road surface marking (lane boundary line 3a, stop line 3b, road sign, etc.), a road boundary (curbstones 3c~3e, guardrails, etc.), and an obstacle (house 3f, fence 3g, object 3h, etc.).

[0035] When the relative positional relationship between the landmark position and the target parking position 2 is stored in the storage device 19, the operation mode of the parking assist device 10 is set to the "landmark object learning mode". Then, the host vehicle 1 is parked at the target parking position 2 by manual driving. In addition, the parking assist device 10 may automatically set the operation mode to the "landmark object learning mode" when the host vehicle 1 is parked at the target parking position 2 by manual driving. When the host vehicle 1 is parked at the target parking position 2 by manual driving, the driver may also select whether to set the operation mode of the parking assist device 10 to the "landmark object learning mode".

[0036] While the vehicle 1 is parked at the target parking position 2 by manual driving, the object sensor 12 detects the position of the marker object around the vehicle 1. The object sensor 12 is a sensor that detects the position of the marker object existing in the detection area within the detection distance range specified by the object sensor 12.

[0037] For example, the parking assist device 10 converts the road marking (in Figure 2 In the example, lane boundary line 3a, stop line 3b), road boundary (in Figure 2 In the example, the grounding part of the curbstones 3c to 3e), obstacles (in Figure 2 In the example, the edges or corners of the house 3f, the wall 3g, the grounding part of the object 3h, etc. are detected as feature points, and the positions of the feature points are used as the marker object positions.

[0038] The parking assist device 10 calculates the feature amount of the detected feature point (e.g., the lightness and attribute of the feature point). The detection of the feature point and the calculation of the feature amount can be performed by methods such as SIFT, SURF, ORB, BRIAK, KAZE, AKAZE, etc. In addition, the parking assist device 10 can also detect the feature points of the point group information obtained by the laser viewfinder, radar, or LiDAR. In this specification, an example of detecting the feature point from the captured image is described. In addition, the feature amount of the feature point does not necessarily need to be detected, and at least the position of the feature point can be detected.

[0039] The parking assistance device 10 obtains the relative positional relationship between the position of the marker object (the position of the feature point) detected by the object sensor 12 and the target parking position 2. For example, the parking assistance device 10 calculates the relative position of the feature point with respect to the host vehicle 1 based on the position of the feature point on the image and the camera information related to the installation state of the camera on the host vehicle 1 (installation position, optical axis angle, and viewing angle). Next, the current position of the host vehicle 1 in the fixed coordinate system at the moment when the position of the marker object is detected by the object sensor 12 is estimated, and the position of the marker object in the fixed coordinate system is calculated based on the estimated current position and the relative position of the marker object with respect to the host vehicle 1. Here, the fixed coordinate system refers to a coordinate system (e.g., a map coordinate system) with a specific location as the coordinate origin.

[0040] The current position of the vehicle 1 in the fixed coordinate system can be estimated, for example, by the positioning device 11, the odometry method and the dead reckoning method. The current position of the vehicle 1 in the fixed coordinate system can also be estimated based on the mapping of the marker object position detected by the object sensor 12 with the known marker object position and high-precision map information.

[0041] Next, the parking assist device 10 determines the target parking position 2 in the fixed coordinate system. For example, the position of the host vehicle 1 when the host vehicle 1 is located at the target parking position 2 can be detected as the target parking position 2. By determining the position of the marker object and the position of the target parking position 2 in the fixed coordinate system, the relative positional relationship between the marker object position and the target parking position 2 is determined.

[0042] The parking assist device 10 stores the relative positional relationship between the marker object position and the target parking position 2 in the storage device 19. For example, the marker object position in the fixed coordinate system and the position of the target parking position 2 may be stored in the storage device 19, respectively. Alternatively, the marker object position in the relative coordinate system with the target parking position 2 as the coordinate origin may be calculated and stored in the storage device 19. Here, an example of a case where the marker object position in the fixed coordinate system and the position of the target parking position 2 are stored in the storage device 19 is described.

[0043] In the following description, the marker object stored in the storage device 19 may be referred to as a “learned marker object.” In addition, the marker object position of the learned marker object may be referred to as a “learned marker object position.” Figure 2 The circular icon indicates the position of the learning completion mark object stored in the storage device 19.

[0044] Next, refer to Figure 3 The parking assistance method by the parking assistance device 10 is described. To utilize parking assistance, the operation mode of the parking assistance device 10 is set to the "parking assistance mode". In the parking assistance mode, the parking assistance device 10 reads the marker object position of the learning completion marker object stored in the storage device 19 and the position of the target parking position 2. Figure 3 The circle icon in represents the marker object position of the learning completed marker object read from the storage device 19 in the parking assist mode.

[0045] The parking assist device 10 detects the relative position of the landmark object around the vehicle 1 with respect to the vehicle 1 through the object sensor 12 as the landmark object position of the landmark object around the vehicle 1. The detection method of the landmark object position is the same as the detection method in the landmark object learning mode. In the parking assist mode, the landmark object position detected by the object sensor 12 is indicated by a triangular icon. Figure 3 In the example of FIG. 1 , lane boundary line 3a, curbstones 3c and 3e, and corners of fence 3g are detected as marker object positions.

[0046] The parking assist device 10 compares each marker object position (triangle icon) detected by the object sensor 12 in the parking assist mode with the learned marker object position (circle icon) read from the storage device 19 to make the same marker object positions correspond to each other. For example, the parking assist device 10 can determine the marker object position with the same or similar feature quantity as the same marker object position. Or, regardless of the feature quantity, the marker object position (triangle icon) detected by the object sensor 12 and the marker object position (circle icon) of the learned marker object read from the storage device 19 are made to correspond to each other by comparing the relative position relationship between the marker object positions (triangle icons) detected by the object sensor 12 and the relative position relationship between the marker object positions (circle icons) of the learned marker object read from the storage device 19. Alternatively, the marker object position (triangle icon) detected by the object sensor 12 and the marker object position (circle icon) of the learned marker object read from the storage device 19 can be made to correspond to each other using both the feature quantity and the relative position relationship of the above-mentioned feature points. Figure 3 In the example of FIG. 1 , the lane boundary line 3 a , the curbstones 3 c and 3 e , and the corners of the fence 3 g correspond.

[0047] The parking assistance device 10 calculates the relative position of the current position of the vehicle 1 relative to the target parking position 2 based on the relative position relationship between the marker object position (triangle icon) detected in the parking assistance mode and the vehicle 1, and the relative position relationship between the learned marker object position (circle icon) corresponding to these feature points (triangle icon) and the target parking position 2.

[0048] For example, the parking assist device 10 may calculate the position of the target parking position 2 on a relative coordinate system (hereinafter referred to as the "vehicle coordinate system") based on the current position of the host vehicle 1. Alternatively, the current position of the host vehicle 1 on the fixed coordinate system may be calculated based on the relative positional relationship between the positions of each marker object (triangle icon) detected in the parking assist mode and the host vehicle 1, and the position of the learning completion marker object (circle icon) on the fixed coordinate system. By determining the position of the host vehicle 1 and the position of the target parking position 2 in the fixed coordinate system, the relative position of the current position of the host vehicle 1 with respect to the target parking position 2 is determined.

[0049] The parking assist device 10 calculates a target driving trajectory from the current position of the host vehicle 1 to the target parking position 2 based on the relative position of the current position of the host vehicle 1 with respect to the target parking position 2. The parking assist device 10 performs parking assist control for assisting the host vehicle 1 to move along the calculated target driving trajectory. For example, the automatic driving of the host vehicle 1 is controlled so as to travel to the target parking position along the calculated target driving trajectory. In addition, for example, the target driving trajectory and the position of the host vehicle 1 are displayed on a display device that can be visually recognized by the user of the host vehicle 1.

[0050] However, in the case where the learning completion marker object position is a marker object position of a movable object (for example, a bicycle placed around the target parking position 2 or a plant planted in a flowerpot), the object may move after the learning completion marker object position is stored. In this case, since the marker object position detected during parking assistance deviates from the learning completion marker object position stored in the storage device 19, or the marker object itself no longer exists, resulting in the marker object position not being detected during parking assistance, if the relative position of the current position of the vehicle 1 relative to the target parking position 2 is calculated based on the learning completion marker object position including these, the calculation accuracy of the relative position may be reduced.

[0051] Therefore, the parking assist device 10 of the embodiment detects the marker object position of the surrounding marker object of the host vehicle 1 at the timing (opportunity) when the host vehicle 1 travels near the target parking position 2 after the learning completion marker object position is stored. In the following description, the marker object position detected at the timing (opportunity) when the host vehicle 1 travels near the target parking position 2 after the learning completion marker object position is stored is sometimes expressed as a "surrounding marker object position".

[0052] The parking assist device 10 counts the number of times the learning completion marker object position stored in the storage device 19 matches the surrounding marker object position for each learning completion marker object position at each opportunity (opportunity) when the host vehicle 1 travels near the target parking position 2 after the learning completion marker object position is stored. Alternatively, the parking assist device 10 may count the number of times the learning completion marker object position stored in the storage device 19 matches the surrounding marker object position for each learning completion marker object position, for each predetermined number of times (e.g., every 2 times, every 3 times) when the host vehicle 1 travels near the target parking position 2 after the learning completion marker object position is stored. Furthermore, the parking assist device 10 may count the number of times the learning completion marker object position stored in the storage device 19 matches the surrounding marker object position at each learning completion marker object position when the host vehicle 1 travels near the target parking position 2 after the learning completion marker object position is stored, and in each predetermined specific scenario (e.g., each time the vehicle leaves the target parking position 2 or each time the vehicle enters the target parking position 2). That is, the parking assist device 10 counts the number of times the learning completion marker object position stored in the storage device 19 matches the surrounding marker object positions for each learning completion marker object position at any time when the vehicle 1 is traveling near the target parking position 2 after storing the learning completion marker object position.

[0053] The parking assist device 10 assigns a higher reliability to a learned marker object position that matches with surrounding marker object positions more times than to a learned marker object position that matches with surrounding marker object positions less times. The parking assist device 10 compares the learned marker object positions having a reliability greater than a predetermined reliability threshold among the learned marker object positions with the positions of marker objects detected around the host vehicle 1, and calculates the relative position of the host vehicle 1 with respect to the target parking position 2.

[0054] Thus, the positions of the fixed object and the movable object are stored as the learning completion mark object positions, and when the movable object moves before the parking assist control is executed later, the learning completion mark object position of the fixed object (immovable object or non-moving object) can be given a higher reliability than the learning completion mark object position of the movable object. As a result, since only the learning completion mark object position of the fixed object can be selected from the learning completion mark object positions and used in the calculation of the relative position of the host vehicle 1 with respect to the target parking position 2, the calculation accuracy of the relative position can be suppressed from being reduced.

[0055] Reference Figure 4 (a)~ Figure 4(d) An example of a method for assigning reliability to the learned marker object position is described. First, in the marker object learning mode, the parking assistance device 10 detects the marker object positions of a plurality of marker objects around the host vehicle 1 when parking the host vehicle 1 at the target parking position 2. The parking assistance device 10 stores the detected positions of the plurality of marker objects as learned marker object positions in the storage device 19.

[0056] Figure 4 The circle icons 30a to 30e in (a) represent the learning completion mark object positions stored in the storage device 19. Figure 4 In the example of (a), the circle icons 30a and 30b indicate the learning completion mark object position of the curb, the circle icon 30c indicates the learning completion mark object position of the lane boundary line, and the circle icon 30d indicates the learning completion mark object position of the house. On the other hand, the circle icon 30e indicates the learning completion mark object position of a movable object 3j (such as a bicycle).

[0057] Reliability is assigned to each of the learned marker object positions 30a to 30e. Immediately after the learned marker object positions 30a to 30e are stored, the reliability of the learned marker object positions 30a to 30e is all the same value (initial value). Figure 4 The lower left value "1" of the learning completion mark object positions 30a to 30e in (a) indicates the reliability of the learning completion mark object positions 30a to 30e. The parking assist device 10 associates the learning completion mark object positions with the reliability and stores them in the storage device 19 as learning completion mark object data 20.

[0058] Reference Figure 4 (b). After storing the learned marker object positions 30a to 30e, the parking assist device 10 detects the marker object positions of the marker objects around the vehicle 1 when the vehicle 1 is traveling near the target parking position 2 (i.e., detects the surrounding marker object positions). The triangular icons 31a to 31e represent the surrounding marker object positions. The triangular icons 31a and 31b represent the surrounding marker object positions of the curb, the triangular icon 31c represents the surrounding marker object positions of the lane boundary line, the triangular icon 31d represents the surrounding marker object positions of the house, and the triangular icon 31e represents the surrounding marker object positions of the movable object 3j.

[0059] The timing for detecting the surrounding marker object positions 31a to 31e may be any timing as long as the host vehicle 1 is traveling near the target parking position 2 after the learned marker object positions 30a to 30e are stored. Figure 4 In the example of (b), the surrounding landmark object positions 31 a to 31 e are detected when the host vehicle 1 is pulled out of the target parking position 2 .

[0060] The timing of the vehicle 1 leaving the target parking position 2 may be, for example, the timing when the driver manually drives the vehicle 1 to leave the target parking position 2. In addition, when the parking assist device 10 has a parking exit assist control function, the timing of detecting the surrounding landmark object positions 31a to 31e may be when the parking exit assist control is executed.

[0061] For example, the exit assist control can be based on the positions of marker objects detected around the vehicle 1 and the positions of marker objects after learning, and calculates the target driving trajectory for the vehicle 1 to exit the parking lot from the current position of the parked vehicle 1, i.e., the target parking position 2, and controls the vehicle 1 to move along the target driving trajectory.

[0062] Alternatively, for example, the timing for detecting the surrounding marker object positions 31 a to 31 e may be when the host vehicle 1 is parked at the target parking position 2 by the parking assist control of the parking assist device 10 after the learned marker object positions 30 a to 30 e are stored.

[0063] Then, each time the host vehicle 1 travels near the target parking position 2, the parking assist device 10 counts the number of times the learning completed marker object positions 30a to 30e coincide with the surrounding marker object positions 31a to 31e for each learning completed marker object position 30a to 30e. For example, when the feature quantities of the learning completed marker object position and the surrounding marker object position are the same or similar to each other, it can be determined that these marker object positions coincide. Figure 4 In the example of (b), the learning completion mark object positions 30a to 30d are respectively consistent with the surrounding mark object positions 31a to 31d. Therefore, the number of times the learning completion mark object positions 30a to 30d are consistent with the surrounding mark object positions is increased by 1. On the other hand, since the movable object 3j moves after the learning completion mark object position 30e is stored, the learning completion mark object position 30e is inconsistent with the surrounding mark object position 31e. Therefore, the number of times the learning completion mark object position 30e is consistent with the surrounding mark object position is not increased.

[0064] The parking assist device 10 assigns higher reliability to the learned marker object positions 30a to 30d having a greater number of coincidences with the surrounding marker object positions than to the learned marker object position 30e having a smaller number of coincidences with the surrounding marker object positions.

[0065] For example, the parking assist device 10 may assign reliability according to the number of times the learned marker object positions 30 a to 30 e coincide with the surrounding marker object positions. Figure 4The lower left value "2" of the learning completion mark object positions 30a to 30d in (b) indicates the reliability of the learning completion mark object positions 30a to 30d, and the lower left value "1" of the learning completion mark object position 30e indicates the reliability of the learning completion mark object position 30e. In this way, the learning completion mark object positions 30a to 30d are given a higher reliability than the learning completion mark object position 30e.

[0066] Reference Figure 4 (c). When the parking assist device 10 performs parking assist control after storing the learned marker object positions 30a to 30e, it detects the marker object positions of the marker objects around the host vehicle 1 (i.e., the surrounding marker object positions). The box icons 32a to 32e represent the surrounding marker object positions detected when the parking assist control is performed. The box icons 32a and 32b represent the surrounding marker object positions of the curb, the box icon 32c represents the surrounding marker object positions of the lane boundary line, the box icon 32d represents the surrounding marker object positions of the house, and the box icon 32e represents the surrounding marker object positions of the movable object 3j.

[0067] The parking assist device 10 selects only the learning completion mark object positions having a reliability greater than a predetermined reliability threshold value among the learning completion mark object positions 30a to 30e. For the sake of explanation, it is assumed that the reliability threshold value is "2". Therefore, only the learning completion mark object positions 30a to 30d of the fixed objects are selected, and the learning completion mark object position 30e of the movable object 3j is not selected.

[0068] The parking assist device 10 calculates the relative position of the current position of the host vehicle 1 with respect to the target parking position 2 based on the selected learning completion marker object positions 30 a to 30 d and the surrounding marker object positions 32 a to 32 d corresponding to the learning completion marker object positions 30 a to 30 d .

[0069] In this way, the parking assist device 10 can select only the learning completion mark object positions 30a to 30d where the fixed object is detected and use them for calculating the relative position of the host vehicle 1 with respect to the target parking position 2, thereby suppressing a decrease in the calculation accuracy of the relative position.

[0070] Reference Figure 4 (d) After the parking assist device 10 has completed the calculation of the relative position of the current position of the host vehicle 1 with respect to the target parking position 2 (for example, after the host vehicle 1 has been moved to the target parking position 2 by the parking assist control), the parking assist device 10 counts the number of times that the learning completion marker object positions 30a to 30e coincide with the surrounding marker object positions 32a to 32e for each learning completion marker object position 30a to 32e.

[0071] exist Figure 4 In the example of (d), the learning completion mark object positions 30a to 30d are respectively consistent with the surrounding mark object positions 32a to 32d. Therefore, the number of times the learning completion mark object positions 30a to 30d are consistent with the surrounding mark object positions is increased by 1. On the other hand, the movable object 3j moves further after detecting the surrounding mark object position 31e, and the learning completion mark object position 30e is inconsistent with the surrounding mark object position 32e. Therefore, the number of times the learning completion mark object position 30e is consistent with the surrounding mark object position is not increased.

[0072] The parking assist device 10 updates the reliability according to the number of times the learning completion marker object positions 30a to 30e match the surrounding marker object positions 32a to 32e. As a result, the reliability of the learning completion marker object positions 30a to 30d increases to "3". On the other hand, the reliability of the learning completion marker object position 30e remains "1".

[0073] By repeatedly detecting the positions of the surrounding objects and updating the reliability, the learning completion mark object position of the fixed object can be given a higher reliability than the learning completion mark object position of the movable object. Therefore, only the learning completion mark object position of the fixed object can be selected from the learning completion mark object positions and used in the calculation of the relative position of the host vehicle 1 with respect to the target parking position 2, thereby suppressing the reduction in the calculation accuracy of the relative position.

[0074] In addition, after storing the learning completion mark object positions 30a to 30e, the parking assisting device 10 may detect the surrounding mark object positions 31a to 31e when the host vehicle 1 travels near the target parking position 2 for the "first time" and assign reliability to the learning completion mark object positions 30a to 30e. For example, the learning completion mark object positions 30a to 30e may be assigned reliability when the host vehicle 1 is pulled out of the target parking position 2 for the first time.

[0075] Therefore, if the movable object 3j moves during the period from the storage of the learning completion mark object positions 30a~30e to the first time the vehicle 1 drives near the target parking position 2, the target driving trajectory is calculated by excluding the learning completion mark object position of the movable object 3j from the first time the vehicle stops at the target parking position 2 after the learning completion mark object position is stored.

[0076] The functional structure of the controller 17 will be described in more detail below. Figure 5The controller 17 functions as an image conversion unit 40, a self-position calculation unit 41, a feature point detection unit 43, a learning completion marker object data generation unit 44, a reliability assignment unit 45, a relative position estimation unit 46, a target trajectory generation unit 47, a steering control unit 48, a vehicle speed control unit 49, and an auxiliary image generation unit 50.

[0077] The image conversion unit 40 converts the image captured by the camera of the object sensor 12 into Figure 2 and Figure 3 Such a bird's-eye view image (surround view monitor image) observed from a virtual viewpoint directly above the host vehicle 1. Hereinafter, the bird's-eye view image converted by the image conversion unit 40 may be referred to as a "surrounding image".

[0078] The own position calculation unit 41 calculates the current position of the host vehicle 1 on the fixed coordinate system based on the dead reckoning method based on the vehicle information output from the vehicle sensor 13. The own position calculation unit 41 can correct the calculated current position through map mapping between the marker object position detected by the object sensor 12 and the known marker object position and high-precision map information.

[0079] The feature point detection unit 43 detects feature points of landmark objects around the host vehicle 1 from the surrounding image output from the image conversion unit 40, and calculates feature values ​​of the feature points. The feature point detection unit 43 outputs the detected feature points and their feature values ​​together with the current position of the host vehicle 1 received from the own position calculation unit 41 to the learning completed landmark object data generation unit 44 and the relative position estimation unit 46.

[0080] The learned marker object data generating unit 44 calculates the positions of the feature points on the fixed coordinate system based on the feature points output from the feature point detecting unit 43 and the current position of the host vehicle 1 .

[0081] In the marker object learning mode, the learning completed marker object data generating unit 44 stores the position of the feature point detected by the feature point detecting unit 43 as the learning completed marker object position in the storage device 19. For example, the learning completed marker object data generating unit 44 sets the reliability of the learning completed marker object position to an initial value, associates the learning completed marker object position with the reliability, and stores the learned completed marker object position in the storage device 19 as the learning completed marker object data 20.

[0082] On the other hand, when the operation mode of the parking assist device 10 is not the landmark object learning mode, the learned landmark object data generating unit 44 outputs feature point information including the position and feature amount information of the feature point to the reliability assigning unit 45 .

[0083] The reliability assigning unit 45 determines whether the host vehicle 1 is traveling near the target parking position 2 . For example, the reliability assigning unit 45 may determine whether the host vehicle 1 is traveling near the target parking position 2 based on the current position of the host vehicle 1 calculated by the own position calculating unit 41 .

[0084] For example, when the host vehicle 1 is being pulled out of the target parking position 2, it can also be determined that the host vehicle 1 is traveling near the target parking position 2. For example, when the distance between the current position of the host vehicle 1 and the target parking position 2 is within a specified value, and the host vehicle 1 is far away from the target parking position 2, or the pull-out assist function is being executed, it can be determined that the host vehicle 1 is being pulled out of the target parking position 2.

[0085] In addition, for example, when the parking assist control for assisting parking at the target parking position 2 is executed, it can be determined that the host vehicle 1 is traveling near the target parking position 2. For example, when the distance between the current position of the host vehicle 1 and the target parking position 2 is within a predetermined value and the operation mode of the parking assist device 10 is the parking assist mode, it can be determined that the parking assist control for assisting parking at the target parking position 2 is being executed.

[0086] When the host vehicle 1 is traveling near the target parking position 2 , the reliability providing unit 45 uses the positions of the feature points output from the learned landmark object data generating unit 44 as the surrounding landmark object positions.

[0087] Each time the vehicle 1 travels near the target parking position 2, the reliability assigning unit 45 counts the number of times the learning completion mark object position stored in the storage device 19 is consistent with the surrounding mark object positions for each learning completion mark object position, and updates the reliability of each learning completion mark object position based on the number of times the learning completion mark object position is consistent with the surrounding mark object positions.

[0088] Specifically, the reliability assigning unit 45 assigns reliability in a manner that the reliability of the learned marker object position that is consistent with the surrounding marker object position more times is higher than the reliability of the learned marker object position that is consistent with the surrounding marker object position less times. For example, the more times the learned marker object position is consistent with the surrounding marker object position, the higher the reliability can be assigned. For example, the number of times the learned marker object position is consistent with the surrounding marker object position can be assigned as reliability.

[0089] The relative position estimation unit 46 reads the learning completion mark object position and its reliability stored in the storage device 19 as the learning completion mark object data 20. Only the learning completion mark object position having a reliability greater than a predetermined reliability threshold value is selected from the read learning completion mark object position. The predetermined reliability threshold value may be, for example, a value obtained by subtracting a predetermined value from the highest reliability value among the reliability values ​​of the learning completion mark object position, or a reliability value from the highest value among the reliability values ​​of the learning completion mark object position to a predetermined rank, or may be set based on an average value of the reliability values ​​of the learning completion mark object position.

[0090] The relative position estimating unit 46 matches the selected learning completed marker object position with the marker object position detected in the parking assist mode, and makes the marker object positions detected for the same object correspond to each other.

[0091] The relative position estimation unit 46 estimates the relative position of the current position of the vehicle 1 relative to the target parking position 2 based on the relative position relationship between the marker object position detected in the parking assistance mode and the vehicle 1, and the relative position relationship between the learned marker object position corresponding to these marker object positions and the target parking position 2.

[0092] For example, the position of the landmark object detected in the parking assistance mode is marked as (x i ,y i ), will be compared with the position of the marked object (x i ,y i ) The corresponding learning completed landmark object positions are marked as (x mi ,y mi )(i=1 to N). For example, the relative position estimation unit 46 may calculate the affine transformation matrix M according to the following equation based on the least squares method: affine .

[0093] (Mathematical formula 1)

[0094]

[0095] in,

[0096]

[0097] The column vector (a1, a2, a3, a4)t may be calculated as shown below using the weighted least squares method.

[0098] (Mathematical formula 2)

[0099] W=diag(w1 … w N )

[0100]

[0101] The relative position estimating unit 46 calculates the position (target x) of the target parking position 2 on the fixed coordinate system included in the parking learning completion sign object data 20 according to the following equation: m , targety m ) is transformed into the position (targetx, targety) of the vehicle coordinate system.

[0102] (Mathematical formula 3)

[0103]

[0104] The target trajectory generation unit 47 generates a target travel trajectory from the current position of the host vehicle 1 on the vehicle coordinate system (ie, the coordinate origin) to the position (targetx, targety) of the target parking position 2 on the vehicle coordinate system.

[0105] The target driving trajectory from the current position of the host vehicle 1 to the target parking position 2 can be calculated by a known method that has been used in a generally known automatic parking device. As an example, the target driving trajectory can be calculated by connecting the current position of the host vehicle 1 to the target parking position 2 with a clothoid curve. In the case where the target driving trajectory includes a turning point, the target driving trajectory can be calculated by connecting the current position of the host vehicle through the turning point to the target parking position 2 with a clothoid curve.

[0106] In addition, the target trajectory generation unit 47 calculates a target vehicle speed curve that sets the moving speed at each position on the target driving trajectory from the current position of the host vehicle to the target parking position 2. For example, the target vehicle speed curve can be calculated based on a predetermined set speed, and a vehicle speed curve can be calculated in which the vehicle 1 accelerates from the current position of the host vehicle 1 to the set speed and then stops at the target parking position 2. In the case where the target driving trajectory includes a turning point, a vehicle speed curve can be calculated in which the vehicle decelerates in front of the turning point and stops at the turning point, then accelerates from the turning point to the set speed, decelerates in front of the target parking position 2, and stops at the target parking position 2. The set speed when calculating the speed curve can also be set based on the curvature of the calculated target driving trajectory so that the speed becomes lower as the curvature increases.

[0107] The steering control unit 48 controls the steering actuator of the actuator 16 so that the host vehicle 1 travels along the target travel trajectory.

[0108] The vehicle speed control unit 49 controls the accelerator actuator and the brake actuator of the actuator 16 so as to change the vehicle speed of the host vehicle 1 according to the moving speed plan calculated by the target trajectory generation unit 47. Thus, the host vehicle 1 is controlled to travel along the target travel trajectory.

[0109] The auxiliary image generation unit 50 generates a parking assistance image showing the target driving trajectory calculated by the target trajectory generation unit 47 and the current position of the host vehicle 1. For example, the parking assistance image may be an image in which the target driving trajectory and the current position of the host vehicle 1 are superimposed on a bird's-eye view or a bird's-eye view of the surroundings of the host vehicle 1 viewed from above. The auxiliary image generation unit 50 displays the parking assistance image on the display device of the human-machine interface 15.

[0110] (action)

[0111] Figure 6 This is a flowchart of an example of processing when storing the marker object position in the marker object learning mode. In step S1, the feature point detection unit 43 detects the marker object position around the target parking position 2 from the surrounding image obtained by photographing the surroundings of the vehicle while the vehicle 1 is parked at the target parking position 2 by manual driving.

[0112] In step S2, the learning completed marker object data generating unit 44 stores the marker object position detected by the feature point detecting unit 43 as the learning completed marker object position in the storage device 19. At this time, the learning completed marker object data generating unit 44 stores the learning completed marker object position and the reliability in the storage device 19 as the learning completed marker object data 20. Then the processing ends.

[0113] Figure 7 This is a flowchart of an example of processing for updating the reliability of the landmark position each time the host vehicle 1 travels near the target parking position 2 after the learning of the landmark position is stored. However, when the host vehicle 1 travels near the target parking position 2, an example of a case where the host vehicle 1 is parked at the target parking position 2 in the parking assist mode is as follows. Figure 8 For example, when the vehicle 1 is pulled out of the target parking position 2, the execution Figure 7 in the processing.

[0114] In step S11 , the feature point detection unit 43 detects surrounding landmark object positions, which are landmark object positions around the host vehicle 1 , from a surrounding image obtained by capturing the surroundings of the host vehicle 1 .

[0115] In step S12, the reliability assigning unit 45 determines whether the learning completed marker object position stored in the storage device 19 is consistent with the marker object position detected in step S11. For the learning marker object position that is consistent with the marker object position detected in step S11, the reliability assigning unit 45 counts the number of times the learning marker object position is consistent with the surrounding marker object positions and adds 1 to the number of times. For the learning completed marker object position that is inconsistent with the marker object position detected in step S11, the number of times the learning completed marker object position is consistent with the surrounding marker object positions is not increased. The reliability assigning unit 45 updates the reliability of each learning completed marker object position according to the number of times the learning completed marker object position is consistent with the surrounding marker object positions. The processing then ends.

[0116] Figure 8 This is a flowchart of an example of parking assist control according to the embodiment. In step S21 , the feature point detection unit 43 detects surrounding landmark object positions, which are landmark object positions around the host vehicle 1 , from a surrounding image obtained by capturing the surroundings of the host vehicle.

[0117] In step S22, the relative position estimation unit 46 estimates the current position of the vehicle 1 relative to the target parking position 2 based on the learning completion mark object position stored in the storage device 19 as the learning completion mark object data 20, which has only a reliability greater than a specified reliability threshold, and the surrounding mark object positions detected in step S21.

[0118] In step S23 , the target trajectory generation unit 479 generates a target travel trajectory and a target vehicle speed curve for causing the host vehicle to travel from the current position of the host vehicle 1 to the target parking position 2 based on the relative position of the host vehicle 1 with respect to the target parking position 2 .

[0119] In step S24, the steering control unit 48 controls the steering angle so that the host vehicle 1 travels along the target travel trajectory. The vehicle speed control unit 49 controls the moving speed of the host vehicle 1 according to the target vehicle speed curve. Thus, the steering control unit 48 and the vehicle speed control unit 49 move the host vehicle 1 to the target parking position 2.

[0120] In step S25, the reliability assigning unit 45 determines whether the learning completed marker object position stored in the storage device 19 is consistent with the marker object position detected in step S21. For the learning completed marker object position that is consistent with the marker object position detected in step S21, the reliability assigning unit 45 counts the number of times the learning completed marker object position is consistent with the surrounding marker object positions and increases the number by 1. For the learning completed marker object position that is inconsistent with the marker object position detected in step S21, the number of times the learning completed marker object position is consistent with the surrounding marker object positions is not increased. The reliability assigning unit 45 updates the reliability of each learning completed marker object position according to the number of times the learning completed marker object position is consistent with the surrounding marker object positions. The processing then ends.

[0121] (Effects of Embodiments)

[0122] (1) The controller 17 stores the positions of multiple marker objects detected around the target parking position 2 when the host vehicle 1 is parked at the target parking position 2 as the learned marker object positions. After storing the learned marker object positions, when the host vehicle 1 is traveling near the target parking position 2, the controller 17 counts the number of times the learned marker object position coincides with the positions of marker objects detected around the host vehicle 1, i.e., the surrounding marker object positions, for each of the multiple learned marker object positions, assigns a higher reliability to the learned marker object position that coincides with the surrounding marker object positions more times than to the learned marker object position that coincides with the surrounding marker object positions less times, compares the learned marker object positions that are above a specified reliability threshold among the learned marker object positions with the positions of marker objects detected around the host vehicle 1, calculates the relative position of the host vehicle 1 with respect to the target parking position 2, calculates the target driving trajectory from the current position of the host vehicle 1 to the target parking position 2 based on the calculated relative position, and executes parking assist control for assisting the host vehicle 1 in moving along the target driving trajectory.

[0123] Thus, when the positions of the fixed object and the movable object are stored as the learning completion mark object position, and the movable object moves before the parking assist control is executed, the learning completion mark object position of the fixed object can be given a higher reliability than the learning completion mark object position of the movable object. Since only the learning completion mark object position of the fixed object can be selected from the learning completion mark object positions and used in the calculation of the relative position of the host vehicle 1 with respect to the target parking position 2, the calculation accuracy of the relative position can be suppressed from being reduced.

[0124] (2) The timing when the vehicle 1 drives near the target parking position 2 after the learning completion mark object position is stored may also include: the timing when the vehicle 1 drives near the target parking position 2 for the first time after the learning completion mark object position is stored. Thus, if a movable object moves after the learning completion mark object position is stored and during the period when the vehicle 1 drives near the target parking position for the first time, the target driving trajectory can be calculated by excluding the learning completion mark object position of the movable object from the time when the vehicle 1 first parks at the target parking position after the learning completion mark object position is stored.

[0125] (3) In addition, the timing when the host vehicle 1 travels near the target parking position 2 after the learning completion mark object position is stored may include when the parking assist control for assisting parking at the target parking position 2 is executed. In this way, the reliability of the learning completion mark object position can be updated every time the parking assist control is executed.

[0126] (4) After the learning completion mark object position is stored, the timing when the host vehicle 1 travels near the target parking position 2 may include the timing when the host vehicle 1 is pulled out of the target parking position 2. Thus, the reliability of the learning completion mark object position can be updated each time the host vehicle 1 is pulled out of the target parking position 2.

[0127] (5) The predetermined reliability threshold may be a value obtained by subtracting a predetermined value from the highest reliability among the reliabilities of the multiple learning completed marker object positions, or may be a reliability from the highest value among the reliabilities of the multiple learning completed marker object positions to a predetermined rank, or may be set based on an average value of the reliabilities of the multiple learning completed marker object positions. In this way, the reliability threshold may be appropriately set.

[0128] (6) Parking assist control is, for example, control of the host vehicle 1 so that the host vehicle 1 moves along the target travel trajectory from the current position of the host vehicle 1 to the target parking position 2. In this way, parking assist can be achieved by controlling the host vehicle 1 so that the host vehicle 1 travels to the target parking position 2.

[0129] (7) The parking assist control may be a control that displays the target driving trajectory and the position of the host vehicle 1 on a display device that can be visually recognized by the user of the host vehicle. Thus, the occupant can visually recognize the target driving trajectory that leads the host vehicle 1 to the target parking position 2 .

[0130] All examples and conditional terms recorded herein are intended for the purpose of teaching in order to help the reader understand the concepts given by the inventor for the present invention and the technical progress, and should be interpreted as not being limited to the above examples and conditions specifically recorded, and the structure of the examples in this specification that represent the superiority and inferiority of the present invention. Although the embodiments of the present invention are described in detail, it should be understood that various changes, substitutions and modifications can be applied thereto without departing from the spirit and scope of the present invention.

[0131] Explanation of symbols

[0132] 40: Image transformation unit, 41: Self-position calculation unit, 43: Feature point detection unit, 44: Learning completion mark object data generation unit, 45: Reliability assignment unit, 46: Relative position estimation unit, 47: Target trajectory generation unit, 48: Steering control unit, 49: Vehicle speed control unit, 50: Auxiliary image generation unit.

Claims

1. A parking assistance method, characterized in that: The positions of a plurality of marker objects detected around the target parking position when the host vehicle is parked at the target parking position are stored as learning completed marker object positions. The number of times the learning completion mark object position is consistent with the position of the mark object detected around the vehicle when the vehicle is traveling near the target parking position after the learning completion mark object position is stored, i.e., the surrounding mark object position, is counted for each of the multiple learning completion mark object positions. The learning completed marker object position having a greater number of coincidences with the surrounding marker object position is given a higher reliability than the learning completed marker object position having a smaller number of coincidences with the surrounding marker object position. The learning completion marker object position that is above a predetermined reliability threshold in the learning completion marker object position is compared with the positions of marker objects detected around the host vehicle to calculate the relative position of the host vehicle with respect to the target parking position, calculating a target driving trajectory from the current position of the host vehicle to the target parking position based on the calculated relative position, A parking assist control is executed to assist the host vehicle to move along the calculated target travel trajectory.

2. The parking assistance method according to claim 1, characterized in that: The timing at which the host vehicle travels near the target parking position after the learning completion marker object position is stored includes the timing at which the host vehicle travels near the target parking position for the first time after the learning completion marker object position is stored.

3. The parking assistance method according to claim 1 or 2, characterized in that: The timing when the host vehicle travels near the target parking position after the learning completion sign object position is stored includes when the parking assist control for assisting parking at the target parking position is executed.

4. The parking assistance method according to any one of claims 1 to 3, characterized in that: The timing at which the host vehicle travels near the target parking position after the learning completion marker object position is stored includes a timing at which the host vehicle is pulled out of the target parking position.

5. The parking assistance method according to any one of claims 1 to 4, characterized in that: The predetermined reliability threshold is set to a value obtained by subtracting a predetermined value from the highest reliability among the reliabilities of the plurality of learned marker object positions.

6. The parking assistance method according to any one of claims 1 to 4, characterized in that: The predetermined reliability threshold is set based on an average value of the reliabilities of the plurality of learning completed marker object positions.

7. The parking assistance method according to any one of claims 1 to 4, characterized in that: The predetermined reliability threshold is set to reliability ranging from the highest value among the reliabilities of the plurality of learning completed marker object positions to a predetermined ranking.

8. The parking assistance method according to any one of claims 1 to 7, characterized in that: The parking assist control is a control of the host vehicle so that the host vehicle moves from the current position of the host vehicle to the target parking position along the target driving trajectory.

9. The parking assistance method according to any one of claims 1 to 8, characterized in that: The parking assist control is a control for displaying the target driving trajectory and the position of the host vehicle on a display device that can be visually recognized by a user of the host vehicle.

10. A parking assist device, characterized in that: have: A sensor for detecting the position of a landmark object around the vehicle, that is, the landmark object position; A controller that stores positions of a plurality of marker objects detected around the target parking position when parking the host vehicle toward the target parking position as learned marker object positions in a prescribed storage device, counts the number of times the learned marker object position coincides with positions of marker objects detected around the host vehicle when the host vehicle is traveling near the target parking position after the learned marker object positions are stored, i.e., surrounding marker object positions, for each of the plurality of learned marker object positions, assigns a higher reliability to the learned marker object position that coincides with the surrounding marker object position more times than to the learned marker object position that coincides with the surrounding marker object position less times, compares the learned marker object positions that are above a prescribed reliability threshold among the learned marker object positions with the positions of marker objects detected around the host vehicle, calculates the relative position of the host vehicle with respect to the target parking position, calculates a target driving trajectory from the current position of the host vehicle to the target parking position based on the calculated relative position, and performs parking assist control for assisting the host vehicle in moving along the calculated target driving trajectory.

Images