Parking assistance method and parking assistance device

Abstract

In this parking assistance method, the own position of a host vehicle is detected (S1), the detection accuracy of the own position is estimated (S2), and the own position of a location where the detection accuracy changes from a state at which the detection accuracy is greater than or equal to a predetermined accuracy to a state at which the detection accuracy is less than the predetermined accuracy is set as a first own position (S13). When the vehicle is parked at the target parking position after the first own position has been set, the relative positional relationship between the target parking position and a marker object present around the target parking position is detected (S10), and when the detection accuracy is less than a predetermined accuracy when the vehicle is parked at the target parking position, the target parking position is parked at the target parking position. The first own position is associated with learned mark object data, which is data indicating the relative position relationship, and stored in a storage device (S11), and when the first own position is associated with the learned mark object data and stored, the learned mark object data is associated with the learned mark object data. The parking of the host vehicle to the target parking position is assisted on the basis of the first host position and the learned marker object data (S40-S43).

Description

Technical Field

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

[0002] Patent Document 1 listed below describes a parking assistance device that detects the own position of a host vehicle based on positioning information and performs parking assistance when it is determined that the own position is close to a target parking position pre-registered in a storage device.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: International Publication No. 2017 / 072941

[0006] Problems to be Solved by the Invention

[0007] In the technology of Patent Document 1, if the detection accuracy of the own position is low, it may not be possible to determine whether the registered target parking position is located near the own vehicle. Summary of the invention

[0008] An object of the present invention is to determine whether or not there is a registered target parking position near a vehicle in a location where the detection accuracy of the vehicle's own position is low.

[0009] In a parking assistance method of one embodiment of the present invention, the self-position of the vehicle is detected, the detection accuracy of the self-position is estimated, and the self-position at a location where the detection accuracy changes from a state above a specified accuracy to a state below the specified accuracy is set as a first self-position. After the first self-position is set, the relative position relationship between the target parking position and the target parking position is detected when the vehicle parks at the target parking position. When the detection accuracy is less than the specified accuracy when the vehicle parks at the target parking position, the first self-position and data representing the relative position relationship, i.e., learning-completed marking object data, are associated and stored in a memory. When the first self-position and the learning-completed marking object data are associated and stored, the vehicle is assisted in parking at the target parking position based on the first self-position and the learning-completed marking object data.

[0010] Effects of the Invention

[0011] According to the present invention, it is possible to determine whether or not there is a registered target parking position near the vehicle in a place where the detection accuracy of the vehicle's own position is low. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a diagram showing a schematic configuration example of a parking assist device.

[0013] Figure 2A It is an explanatory diagram of an example of a process of registering a target parking position.

[0014] Figure 2B It is an explanatory diagram of an example of processing when performing parking assistance.

[0015] Figure 3 yes Figure 1 A block diagram showing an example of the functional structure of a controller in FIG.

[0016] Figure 4A This is a schematic diagram showing a situation where the detection accuracy of the own position decreases.

[0017] Figure 4B This is a schematic diagram showing a situation where the detection accuracy of the own position decreases.

[0018] Figure 5A It is an explanatory diagram of a process when performing parking assistance to a target parking position.

[0019] Figure 5B It is an explanatory diagram of a process when performing parking assistance to a target parking position.

[0020] Figure 6 This is a flowchart of an example of storage processing of learning completed landmark object data.

[0021] Figure 7 This is a flowchart of an example of processing when performing parking assistance. DETAILED DESCRIPTION

[0022] Reference Figure 1 . The vehicle 1 is provided with a parking assistance device 10 for assisting the vehicle 1 in parking at a target parking position. The parking assistance device 10 assists the vehicle 1 in traveling along a target driving trajectory from the current position to the target parking position. For example, an automatic driving operation may be performed to control the vehicle 1 so that the vehicle 1 travels along the target driving trajectory of the vehicle 1 to the target parking position (i.e., controlling all or part of the steering angle, driving force, and braking force of the vehicle 1 to automatically implement all or part of the control of the vehicle 1 traveling along the target driving trajectory). The parking of the vehicle 1 may also be assisted by displaying the target driving trajectory and the current position of the vehicle 1 on a display device that can be visually recognized by the occupants of the vehicle 1.

[0023] The positioning device 11 measures the current position of the vehicle 1, that is, its own position. The positioning device 11 may have a global navigation satellite system (GNSS) receiver such as a global positioning system (GPS) receiver, or may receive position information from a wireless local area network (LAN) access point or a mobile phone base station to measure its own position. Map data is stored in the map database (map DB) 12. The map data stored in the map DB 12 may be, for example, high-precision map data suitable as a map for autonomous driving. The human-machine interface (HMI) 13 is an interface device for sending and receiving information between the parking assistance device 10 and the occupants. The shift switch (shift SW) 14 is a switch used by the occupants of the vehicle 1 (such as the driver) or the parking assistance device 10 to switch the gear of the vehicle 1.

[0024] The external sensor 15 detects objects within a specified distance range from the vehicle 1. The external sensor 15 detects the surrounding environment of the vehicle 1, such as the relative position between the vehicle 1 and the objects, the distance between the vehicle 1 and the objects, and the direction in which the objects exist. The external sensor 15 may include, for example, a camera that captures the surrounding environment of the vehicle 1. In the following description, the camera possessed by the external sensor 15 is referred to as a "camera". The external sensor 15 may include a distance measuring device such as a laser viewfinder, a radar, and a LiDAR. The vehicle sensor 16 detects various information (vehicle information) of the vehicle 1. The vehicle sensor 16 may include, for example, a vehicle speed sensor, a wheel speed sensor, a three-axis acceleration sensor, a steering angle sensor, a steering angle sensor, a gyro sensor, and a yaw rate sensor.

[0025] The controller 17 is an electronic control unit that performs parking assist control. The controller 17 includes peripheral components such as a processor 20 and a storage device 21. The processor 20 may be, for example, a CPU or an MPU. The storage device 21 may include a semiconductor memory, a magnetic memory, an optical memory, etc. The functions of the controller 17 are realized, for example, by causing the processor 20 to execute a computer program stored in the storage device 21. The steering gear actuator 19a controls the steering direction and steering amount of the steering mechanism of the vehicle 1 according to the control signal of the controller 17. The accelerator actuator 19b controls the accelerator opening of the engine or the drive motor, i.e., the drive device, according to the control signal of the controller 17. The brake actuator 19c activates the brake device according to the control signal of the controller 17.

[0026] Next, the parking assistance control of the parking assistance device 10 will be described. When the parking assistance of the parking assistance device 10 is used, the target parking position where the host vehicle 1 is to be parked is registered in the parking assistance device 10. Specifically, the marker objects existing around the target parking position are extracted and stored in advance in the storage device 21. In the following description, the marker objects around the target parking position stored in the storage device 21 are expressed as "learning completion marker objects". Figure 2A 1 is an explanatory diagram of an example of a process of registering a target parking position, and a circle icon indicates a learning completion mark object. When registering the target parking position 30 in the parking assist device 10, the occupant performs an operation (hereinafter sometimes referred to as a "registration operation") to instruct registration of the target parking position 30. The registration operation may be, for example, an operation of a "parking position registration switch" prepared on the HMI 13.

[0027] For example, when the host vehicle 1 is located near the target parking position 30 (for example, when the occupant parks the host vehicle 1 at the target parking position 30 by manual driving), the parking assist device 10 detects the marker object around the host vehicle 1 by the external sensor 15 and stores it as a learning completion marker object. For example, the marker object can be detected from the surrounding image obtained by capturing the surrounding of the host vehicle 1 by a camera. Alternatively, the marker object around the host vehicle 1 can be detected by a distance measuring device. The parking assist device 10 stores the learning completion marker object data related to the learning completion marker object in the storage device 21. For example, the learning completion marker object data includes data indicating the feature quantity of the learning completion marker object (hereinafter referred to as "feature quantity data"), data on the relative position relationship between the learning completion marker object and the target parking position (hereinafter referred to as "relative position data"), and coordinate data of the target parking position 30 in a coordinate system (hereinafter referred to as "map coordinate system") with a fixed point as a reference point (hereinafter referred to as "target parking position coordinate data").

[0028] As relative position data, for example, the relative position of a learning completion mark object with respect to the target parking position 30 may be stored. For example, the parking assist device 10 can obtain the position of the learning completion mark object detected when the host vehicle 1 is located at the target parking position 30 as the relative position of the learning completion mark object with respect to the target parking position 30. The coordinates of the learning completion mark object and the target parking position 30 in the map coordinate system may also be stored. As target parking position coordinate data, the own position of the host vehicle 1 when the host vehicle 1 is located at the target parking position 30 may be stored. At this time, the own position of the host vehicle 1 measured by the positioning device 11 may be stored when the detection accuracy of the positioning device 11 is greater than the prescribed accuracy, and the own position of the host vehicle 1 estimated by an odometry method such as a dead reckoning method may be stored when the detection accuracy of the positioning device 11 is less than the prescribed accuracy.

[0029] Figure 2B 1 is an explanatory diagram of an example of processing when performing parking assistance. The parking assistance device 10 starts parking assistance for the host vehicle 1 when the host vehicle 1 is located near the registered target parking position 30 and an operation (hereinafter sometimes referred to as "start operation") of the occupant is performed to instruct the start of parking assistance control for the host vehicle 1 to move to the target parking position 30. The start operation may be, for example, an operation of a "parking assistance start switch" prepared on the HMI 13, or a shift operation for switching between forward and reverse movement of the host vehicle 1.

[0030] The parking assist device 10 extracts landmark objects around the host vehicle 1 using the external sensor 15. In the following description, the landmark objects around the host vehicle 1 extracted when performing parking assistance are referred to as "surrounding landmark objects". Figure 2B , the triangular icon represents a surrounding marker object. The parking assist device 10 matches the learning completion marker object with the surrounding marker objects and makes the same feature points correspond to each other. Then, based on the relative positional relationship between the surrounding marker objects detected when the parking assist is implemented and the host vehicle 1, and the relative positional relationship between the learning completion marker object corresponding to the surrounding marker object and the target parking position 30, the relative position of the host vehicle 1 with respect to the target parking position 30 is calculated. For example, the parking assist device 10 calculates the position of the target parking position 30 on a coordinate system (hereinafter referred to as "vehicle coordinate system") based on the current position of the host vehicle 1. In addition, when the coordinates of the learning completion marker object and the target parking position 30 on the map coordinate system are stored in the storage device 21, the coordinates of the target parking position 30 on the map coordinate system may be converted into coordinates on the vehicle coordinate system based on the positions of the surrounding marker objects detected when the parking assist is implemented and the positions of the learning completion marker objects on the map coordinate system. The position of the vehicle 1 on the map coordinate system can also be determined based on the positions of surrounding marker objects detected when implementing parking assistance and the positions of learning completion marker objects in the map coordinate system. The relative position of the vehicle 1 relative to the target parking position 30 can be calculated based on the difference between the coordinates of the vehicle 1 in the map coordinate system and the coordinates of the target parking position 30.

[0031] The parking assist device 10 calculates a target driving trajectory 33 from the current position 32 of the host vehicle 1 to the target parking position 30 based on the relative position of the host vehicle 1 to the target parking position 30. The parking assist device 10 performs parking assist control of the host vehicle 1 based on the calculated target driving trajectory 33.

[0032] The functional structure of the controller 17 is described in more detail below. Figure 3. When the HMI control unit 40 detects the registration operation of the occupant for the target parking position 30, the HMI control unit 40 outputs a map generation instruction to the map generation unit 45 to store the learning completion mark object data in the storage device 21. When the start operation of the parking assist control for the registered target parking position 30 is detected, a control start instruction to start the parking assist control for the target parking position 30 is output to the parking assist control unit 41. The image conversion unit 42 converts the camera image into a bird's-eye view image observed from a virtual viewpoint directly above the host vehicle 1. The image conversion unit 42 converts the camera image into a bird's-eye view image at a predetermined interval, and accumulates the converted bird's-eye view images along the driving path of the host vehicle 1, thereby generating an image of the surrounding area of ​​the host vehicle 1, that is, a surrounding image.

[0033] The own position calculation unit 43 calculates the current position of the own vehicle 1 on the map coordinate system as the own position by the odometry method (e.g., dead reckoning method) based on the vehicle information output from the vehicle sensor 16. The own position calculation unit 43 corrects the calculation result of the own position based on the detection result of the own position of the positioning device 11. The own position calculation unit 43 estimates the detection accuracy of the positioning device 11. For example, when the positioning device 11 is equipped with a GNSS receiver, the own position calculation unit 43 can estimate the detection accuracy of the positioning device 11 based on the number of navigation satellites captured. When the positioning device 11 performs positioning based on the position information from the wireless LAN access point or the mobile phone base station, the detection accuracy of the positioning device 11 can be estimated based on the number of wireless LAN access points or mobile phone base stations captured or the reception strength of the position information. The map generation unit 45 and the docking unit (corresponding unit) 47 described later can also estimate the detection accuracy of the positioning device 11 using the same method.

[0034] The parking assistance device 10 may also detect the own position of the host vehicle 1 using high-precision map data instead of or in addition to the positioning device 11. For example, the parking assistance device 10 may also detect the landmark objects around the host vehicle 1 through the external sensor 15, and detect the own position of the host vehicle 1 through the mapping match between the detected landmark objects and the high-precision map data. In this case, the detection accuracy of the positioning based on the high-precision map data can be estimated based on the accuracy of the mapping match (e.g., matching error).

[0035] The marker object detection unit 44 detects the marker object based on the surrounding image output from the image conversion unit 42. The marker object detection unit 44 can detect the position of the feature point of the marker object and its image feature value as the marker object. The marker object detection unit 44 outputs the position of the detected feature point and the image feature value as the marker object data to the map generation unit 45 and the docking unit 47. In addition, synchronously with the detection of the marker object, the own position obtained from the own position calculation unit 43 is output to the map generation unit 45 and the docking unit 47.

[0036] When receiving a map generation instruction from the HMI control unit 40 (i.e., when a registration operation of the target parking position 30 is performed), the map generation unit 45 generates learning-completed landmark object data and stores it in the storage device 21 as map data 46. For example, the map generation unit 45 receives landmark object data and the own position of the host vehicle 1 on the map coordinate system synchronized with the landmark object data from the landmark object detection unit 44. The map generation unit 45 obtains the position information of the target parking position 30 in the map coordinate system. It is also possible to obtain the own position calculated by the own position calculation unit 43 when the host vehicle 1 is located at the target parking position 30 as the position information of the target parking position 30. The map generation unit 45 generates relative position data based on the position of the feature point included in the landmark object data, the position information of the host vehicle 1 synchronized therewith, and the position information of the target parking position 30. In addition, the map generation unit 45 obtains feature amount data from the landmark object data output from the landmark object detection unit 44. The position information of the target parking position 30 is used as the target parking position coordinate data. The learning completion sign object data including the relative position data, feature amount data, and target parking position coordinate data are stored in the storage device 21 as map data 46 .

[0037] When receiving a control start instruction from the HMI control unit 40, the parking assist control unit 41 outputs a parking position calculation instruction to the docking unit 47. The docking unit 47 receives the marker object data output from the marker object detection unit 44 as the marker object data of the surrounding marker objects, and synchronously receives the own position of the vehicle 1 in the map coordinate system. The docking unit 47 determines whether the vehicle 1 is located near the registered target parking position 30 based on the target parking position coordinate data included in the learned marker object data stored in the storage device 21. In addition, the docking unit 47 can also compare the feature quantity data included in the learned marker object data stored in the storage device 21 with the marker object data of the marker objects around the vehicle 1 output from the marker object detection unit 44 to determine whether the vehicle 1 is located within the range where a marker object that can match the learned marker object data near the target parking position 30 can be detected.

[0038] When the vehicle 1 is located near the registered target parking position 30, or when the vehicle 1 is located within a range where a marker object that can match the learning completed marker object data can be detected, the docking unit 47 matches the learning completed marker object with the surrounding marker objects and makes the marker objects with the same feature points correspond to each other. The docking unit 47 calculates the relative position of the vehicle 1 with respect to the target parking position 30 based on the relative position relationship between the surrounding marker objects and the vehicle 1, and the relative position relationship between the learning completed marker objects corresponding to the surrounding marker objects and the target parking position 30. For example, the surrounding marker objects are expressed as (x i ,y i ), will be combined with surrounding landmark objects (x i ,y i ) The corresponding learning completion mark object is expressed as (x mi ,y mi )(i=1 to N). The docking unit 47 calculates the affine transformation matrix M according to the following equation based on the least squares method: affine .

[0039] Equation 1

[0040]

[0041] in,

[0042]

[0043] The docking unit 47 converts the position (targetx) of the target parking position 30 on the map coordinate system stored in the map data 46 into the following equation: m , targety m ) is transformed into the position (targetx, targety) of the vehicle coordinate system.

[0044] Equation 2

[0045]

[0046] The target trajectory generation unit 48 calculates a target driving trajectory and a target vehicle speed curve from the current position of the host vehicle 1 on the vehicle coordinate system to the target parking position 30. The steering control unit 49a controls the steering actuator 19a so that the host vehicle 1 travels along the target driving trajectory. The vehicle speed control unit 49b controls the accelerator actuator 19b and the brake actuator 19c so that the vehicle speed of the host vehicle 1 changes according to the target vehicle speed curve. When the host vehicle 1 reaches the target parking position 30 and the parking assist control is completed, the parking assist control unit 41 activates the parking brake 18 and switches the gear to the parking gear (P gear).

[0047] Next, the processing when the detection accuracy of the own position decreases is described. For example, when the positioning device 11 detects the own position of the vehicle 1 using a GNSS receiver, a wireless LAN access point, etc., the detection accuracy of the own position may decrease in an indoor parking lot (such as an underground parking lot). When the own position is detected using high-precision map data, the detection accuracy of the own position may decrease in places where the high-precision map data is not fully complete.

[0048] Figure 4A and Figure 4B FIG. 5 is a schematic diagram showing a state where the target parking position 30 is stored in the indoor parking lot 50. Figure 4A In the example of FIG. 5 , the target parking position 30 in the parking frame 53n for parking along the relatively short driving trajectory 52S after entering the indoor parking lot 50 from the entrance 51 is stored. Figure 4B In the example of FIG. 5 , the target parking position 30 in the parking frame 53f for parking along the relatively long driving trajectory 52L is stored.

[0049] In the case where the detection accuracy of the own position by the positioning device 11 and the high-precision map data is low, the own position calculation unit 43 cannot correct the own position calculated by the odometry method. Therefore, when the travel distance of the own vehicle 1 increases, errors will accumulate in the calculation result of the own position, and the position accuracy of the target parking position 30 stored as the target parking position coordinate data after learning the landmark object data will decrease. In addition, the position accuracy of the own position at the time of executing the parking assist control will also decrease. As a result, when executing the parking assist control, the docking unit 47 may not be able to determine whether the own vehicle 1 is located near the registered target parking position 30.

[0050] Therefore, when the map generation unit 45 determines that the detection accuracy of the self-position has changed from a state of being greater than or equal to the prescribed accuracy to a state of being less than the prescribed accuracy, the self-position detected at the location where the detection accuracy has changed to a state of being less than the prescribed accuracy may be set as the first self-position P1. For example, the self-position detected before the detection accuracy of the self-position has changed from a state of being greater than or equal to the prescribed accuracy to a state of being less than the prescribed accuracy may be set as the first self-position P1. For example, the self-position detected before the detection accuracy of the self-position has changed from a state of being greater than or equal to the prescribed accuracy to a state of being less than the prescribed accuracy may be temporarily stored as the first self-position P1.

[0051] The map generation unit 45 determines whether to generate and store learning completion marker object data in the storage device 21 after setting the first self-position P1. For example, when the detection accuracy is less than the prescribed accuracy when the host vehicle 1 is parked at the target parking position 30, the map generation unit 45 may associate the information of the first self-position P1 with the learning completion marker object data and store them in the storage device 21. For example, the map generation unit 45 may also determine whether the occupant has performed a registration operation in a state where the detection accuracy is less than the prescribed accuracy. In the case where it is determined that the learning completion marker object data is generated and stored in the storage device 21 in a state where the detection accuracy is less than the prescribed accuracy, the map generation unit 45 associates the information of the first self-position P1 with the learning completion marker object data and stores them in the storage device 21.

[0052] For example, the map generation unit 45 may determine whether to generate learning completion marker object data and store it in the storage device 21 when the detection accuracy of the first self-position P1 changes from a state of more than a prescribed accuracy to a state of less than a prescribed accuracy and the state of less than a prescribed accuracy continues. For example, the map generation unit 45 may determine whether the occupant has performed a registration operation when the state of less than a prescribed accuracy continues. When it is determined that the learning completion marker object data is generated and stored in the storage device 21 when the state of less than a prescribed accuracy continues, the map generation unit 45 may associate the information of the first self-position P1 with the learning completion marker object data and store them in the storage device 21.

[0053] Thus, the learning completion sign object data of the target parking position 30 registered in a state where the detection accuracy is less than the prescribed accuracy can be associated with the position information (first own position P1) detected with an accuracy greater than or equal to the prescribed accuracy and stored in the storage device 21. As a result, even when the detection accuracy of the own position is less than the prescribed accuracy, by comparing whether the own position of the own vehicle 1 is near the first own position P1, it is possible to determine whether there is a registered target parking position 30 near the own position of the own vehicle 1, and read the learning completion sign object data from the storage device 21. For example, the detection accuracy of the own position is temporarily stored at a point where the detection accuracy changes from a state where the detection accuracy is greater than or equal to the prescribed accuracy to a state where the detection accuracy is less than the prescribed accuracy. When the parking assist control is started, the learning completion sign object data of the registered target parking position 30 can be read by comparing the stored point with the first own position P1. In addition, for example, when a plurality of target parking positions 30 are registered in the storage device 21, the target parking position 30 registered at a place where the detection accuracy of the own position is less than the prescribed accuracy can be correctly selected.

[0054] Figure 5A and Figure 5B1 is a schematic diagram showing a state in which parking assist control is performed toward the target parking position 30 at a time point after the target parking position 30 is registered. The positioning device 11 detects the own position (sometimes referred to as "second own position" in the following description) P2 of the host vehicle 1 at a time point after the target parking position 30 is registered. The second own position P2 may be detected based on high-precision map data instead of or in addition to the positioning device 11.

[0055] Figure 5A and Figure 5B The range 54 indicated by the dotted line in the figure indicates a range where the distance from the first self position P1 is less than the first threshold value Dt1. When the distance between the second self position P2 and the first self position P1 becomes less than the first threshold value Dt1, the docking unit 47 reads the learning completion mark object data associated with and stored in the first self position P1 from the storage device 21. Thus, the learning completion mark object data of the target parking position 30 registered in the indoor parking lot 50 can be read from the storage device 21 to check whether the self position of the host vehicle 1 is located near the first self position P1.

[0056] For example, the docking unit 47 may estimate the detection accuracy of the second self-position P2 and determine whether the detection accuracy has changed from a state of being greater than a predetermined accuracy to a state of being less than a predetermined accuracy. The docking unit 47 may also read the learning completion mark object data associated with the first self-position P1 and stored from the storage device 21 when the distance between the second self-position P2 detected before the detection accuracy changes to an accuracy less than the predetermined accuracy and the first self-position P1 becomes less than the first threshold value Dt1.

[0057] For example, when the detection accuracy of the second self position P2 changes from a state of being greater than a predetermined accuracy to a state of being less than a predetermined accuracy, the docking unit 47 may determine whether the first self position P1 stored in the storage device 21 exists within a range of the first threshold value Dt1 from the second self position P2 detected before the detection accuracy changes to being less than a predetermined accuracy. That is, the first self position P1 existing within a range of the first threshold value Dt1 from the second self position P2 may be determined.

[0058] Then, at the time point when the parking position calculation instruction is received (ie, the time point when the start operation is received), the docking unit 47 can read out the learning completion mark object data associated with the determined first self position P1 and stored from the storage device 21 .

[0059] For example, the docking unit 47 may temporarily store the second self position P2 when the detection accuracy of the second self position P2 changes from a state of being greater than or equal to a state of being less than the specified accuracy and the distance between the second self position P2 detected before the detection accuracy changes to being less than the specified accuracy and the first self position P1 becomes the first threshold value Dt1. Then, at the time of receiving the parking position calculation instruction, the docking unit 47 may read the learning completion mark object data associated with the first self position P1 that exists within the range of the temporarily stored second self position P2 by the first threshold value Dt1 and stored from the storage device 21.

[0060] As described above, the self-position calculated by the self-position calculation unit 43 using the odometry method accumulates calculation errors as the travel distance of the host vehicle 1 increases. Therefore, even if the detection accuracy of the self-position of the positioning device 11 and the high-precision map data is less than the specified accuracy and the calculation result of the self-position cannot be corrected, it can be expected that the position accuracy of the self-position calculated by the self-position calculation unit 43 is high before the travel distance of the host vehicle 1 increases. For example, Figure 5A In the case of , the travel distance of the host vehicle 1 from when the detection accuracy of the own position becomes less than the specified accuracy to when it approaches the registered target parking position 30 is short. Therefore, it is possible to accurately determine whether the host vehicle 1 is sufficiently close to the target parking position 30 (i.e., whether the docking unit 47 can match the surrounding landmark objects with the learning completion landmark objects of the target parking position 30). Figure 5B In the case of , the travel distance until approaching the target parking position 30 becomes longer, and the position accuracy of the own position calculated by the own position calculation unit 43 decreases. Therefore, it cannot be determined whether the own vehicle 1 has sufficiently approached the target parking position 30.

[0061] Therefore, the docking unit 47 can calculate the travel distance of the host vehicle 1 after the detection accuracy of the second self-position P2 changes from a state of being greater than the prescribed accuracy to a state of being less than the prescribed accuracy. When the calculated travel distance is less than the second threshold value Dt2, the docking unit 47 can read the learning completion marker object data from the storage device 21 based on the self-position of the host vehicle 1 calculated by the self-position calculation unit 43 according to the odometry method. For example, the docking unit 47 can compare the self-position of the host vehicle 1 calculated according to the odometry method with the target parking position coordinate data of the learning completion marker object data stored in the storage device 21, and read the learning completion marker object data of the target parking position 30 near the self-position of the host vehicle 1 from the storage device 21.

[0062] For example, the docking unit 47 determines whether the host vehicle 1 is located near the registered target parking position 30 by comparing the own position of the host vehicle 1 calculated by the own position calculation unit 43 according to the odometry method with the target parking position coordinate data of the learning completion mark object data stored in the storage device 21. When the host vehicle 1 is located near the registered target parking position 30 and receives the start operation of the occupant, the learning completion mark object data of the target parking position 30 near the host vehicle 1 is read from the storage device 21 and the parking assistance of the host vehicle 1 is implemented.

[0063] On the other hand, when the driving distance after the detection accuracy of the second self-position P2 becomes less than the specified accuracy exceeds the second threshold value Dt2, the learned marker object data of the marker object having feature quantities similar to those of the surrounding marker objects around the vehicle 1 can be read from the storage device 21, which is associated with and stored in the first self-position P1 within the distance range below the first threshold value Dt1 from the second self-position P2 detected before the detection accuracy changes to an accuracy less than the specified accuracy.

[0064] Specifically, the docking unit 47 compares the feature quantity data of the learning completion marker object 57 of the learning completion marker object data of the target parking position 30 associated with the first own position P1 located in the distance range less than the first threshold value Dt1 from the second own position P2 and stored in the storage device 21, and the feature quantity of the surrounding marker object 56 detected around the own vehicle 1 when the parking assist is implemented. Based on whether the feature quantity of the surrounding marker object is similar to the feature quantity data, it is determined whether the own vehicle 1 is located near the target parking position 30. In the case where there are a plurality of target parking positions 30 associated with the first own position P1 located in the distance range less than the first threshold value Dt1 from the second own position P2 and stored in the storage device 21, it is determined whether the own vehicle 1 is located near the target parking position 30 for all of these plurality of target parking positions 30.

[0065] When it is determined that the host vehicle 1 is located near the target parking position 30 and a start operation by the occupant is received, the learning completion marker object data of the target parking position 30 near the host vehicle 1 is read from the storage device 21 and parking assistance for the host vehicle 1 is performed.

[0066] Reference Figure 6. In step S1, the positioning device 11 detects its own position. In step S2, the map generation unit 45 estimates the detection accuracy of its own position. When the detection accuracy is less than the prescribed accuracy (step S3: Yes), the process proceeds to step S4. When the detection accuracy is greater than the prescribed accuracy (step S3: No), the process proceeds to step S7. In step S4, the map generation unit 45 determines whether the value of the first flag FLG1 is True. The first flag FLG1 is a determination value set to True or False in a prescribed step on the flowchart. For example, the value of FLG1 is False in step S6, and the value of FLG1 is True in step S7. When the value of the first flag FLG1 is True (step S4: Yes), the process proceeds to step S5. When the value of the first flag FLG1 is False (step S4: No), the process proceeds to step S9.

[0067] In step S5, the map generation unit 45 temporarily stores the self-position detected in step S1 as the first self-position P1. In step S6, the map generation unit 45 sets the value of the first flag FLG1 to False. Then, the process proceeds to step S9. In step S7, the map generation unit 45 sets the value of the first flag FLG1 to True. In step S8, the map generation unit 45 deletes the first self-position P1 stored in step S5. Then, the process proceeds to step S9.

[0068] In step S9, the HMI control unit 40 determines whether the occupant has performed a registration operation. If the occupant has performed a registration operation (step S9: Yes), the process proceeds to step S10. If the occupant has not performed a registration operation (step S9: No), the process returns to step S1. In step S10, the landmark object detection unit 44 detects landmark objects around the target parking position 30. In step S11, the map generation unit 45 generates learning completion landmark object data and stores it in the storage device 21. In step S12, the map generation unit 45 determines whether the value of the first flag FLG1 is False. If the value of the first flag FLG1 is False (step S12: Yes), the process proceeds to step S13. If the value of the first flag FLG1 is True (step S12: No), the process ends. In step S13, the map generation unit 45 associates the information of the first self position P1 temporarily stored in step S5 with the learning completion landmark object data stored in step S11 and stores them in the storage device 21. Then, the process ends.

[0069] Reference Figure 7. In step S20, the positioning device 11 detects the second self-position P2. In step S21, the docking unit 47 estimates the detection accuracy of the second self-position P2. When the detection accuracy is less than the prescribed accuracy (step S22: Yes), the process proceeds to step S23. When the detection accuracy is greater than the prescribed accuracy (step S22: No), the process proceeds to step S29. In step S23, the docking unit 47 determines whether the value of the second flag FLG2 is True. The second flag FLG2 is a determination value set to True or False in a prescribed step on the flowchart. For example, the value of FLG2 is False in step S28, and the value of FLG2 is True in step S29. When the value of the second flag FLG2 is True (step S23: Yes), the process proceeds to step S24. When the value of the second flag FLG2 is False (step S23: No), the process proceeds to step S32.

[0070] In step S24, the docking unit 47 determines whether the distance between the second self position P2 and the first self position P1 is less than the first threshold value Dt1. When the distance between the second self position P2 and the first self position P1 is less than the first threshold value Dt1 (step S24: Yes), the process proceeds to step S25. When the distance between the second self position P2 and the first self position P1 is not less than the first threshold value Dt1 (step S24: No), the process proceeds to step S27. In step S25, the docking unit 47 temporarily stores the second self position P2. In step S26, the docking unit 47 sets the value of the third flag FLG3 to True. The third flag FLG3 is a determination value set to True or False in a predetermined step on the flowchart. For example, the value of FLG3 in step S26 is True, and the value of FLG3 in steps S27 and S31 is False. Then, the process proceeds to step S28. In step S27, the docking unit 47 sets the value of the third flag FLG3 to False. Then, the process proceeds to step S28. In step S28, the docking unit 47 sets the value of the second flag FLG2 to False. Then, the process proceeds to step S32.

[0071] In step S29, the docking unit 47 sets the value of the second flag FLG2 to True. In step S30, the docking unit 47 deletes the second self position P2 previously stored in step S25. In step S31, the docking unit 47 sets the value of the third flag FLG3 to False. Then, the process proceeds to step S32.

[0072] In step S32, the marker object detection unit 44 detects surrounding marker objects existing around the vehicle 1. In step S33, the HMI control unit 40 determines whether the occupant has performed a start operation. If the occupant has performed a start operation (step S33: Yes), the process proceeds to step S34. If the occupant has not performed a start operation (step S33: No), the process returns to step S20.

[0073] In step S34, the docking unit 47 determines whether the value of the second flag FLG2 is False. If the value of the second flag FLG2 is False (step S34: Yes), the process proceeds to step S36. If the value of the second flag FLG2 is True (step S34: No), the process proceeds to step S35. In step S35, the docking unit 47 reads the learning completion mark object data of the registered target parking position 30 near the second own position P2 as the current position of the host vehicle 1 from the storage device 21. Then, the process proceeds to step S40.

[0074] In step S36, the docking unit 47 determines whether the value of the third flag FLG3 is True. If the value of the third flag FLG3 is True (step S36: Yes), the process proceeds to step S37. If the value of the third flag FLG3 is False (step S36: No), the process ends.

[0075] In step S37, the docking unit 47 determines whether the travel distance after the detection accuracy of the second self-position P2 becomes less than the prescribed accuracy is less than the second threshold value Dt2. When the travel distance is less than the second threshold value Dt2 (step S37: Yes), the process proceeds to step S38. When the travel distance is not less than the second threshold value Dt2 (step S37: No), the process proceeds to step S39. In step S38, the docking unit 47 reads the learning completion marker object data from the storage device 21 based on the self-position calculated by the self-position calculation unit 43 by the odometry method. Then, the process proceeds to step S40. In step S39, the docking unit 47 reads the learning completion marker object data of the marker object having a feature value similar to that of the surrounding marker objects from the learning completion marker object data associated with the first self-position P1 within the distance range within which the second self-position P2 stored in step S25 is less than the first threshold value Dt1 and stored from the storage device 21. Then, the process proceeds to step S40.

[0076] In step S40, the docking unit 47 matches the surrounding marker objects with the learning completion marker object data to calculate the relative position of the vehicle 1 relative to the target parking position 30. In step S41, the target trajectory generation unit 48 calculates the target driving trajectory and the target vehicle speed curve. In step S42, the steering control unit 49a and the vehicle speed control unit 49b control the steering actuator 19a, the accelerator actuator 19b and the brake actuator 19c based on the target driving trajectory and the target vehicle speed curve. In step S43, the parking assist control unit 41 activates the parking brake 18 and switches the gear to the P gear when the parking assist control is completed. Then, the process ends.

[0077] (Effects of Embodiments)

[0078] (1) According to the parking assist method described in the first aspect, even in a place where the detection accuracy of the own position is less than the predetermined accuracy, it is possible to determine whether there is a registered target parking position near the own vehicle 1 and read the learning completion marker object data from the storage device.

[0079] According to the parking assistance method of the second aspect, even when the detection accuracy of the own position is less than the predetermined accuracy, the learning completion marker object data of the registered target parking position can be used to assist the host vehicle 1 in parking the target parking position.

[0080] According to the parking assistance method of the third aspect, even when the detection accuracy of the own position is less than the predetermined accuracy, the learning completion mark object data of the registered target parking position can be used to assist the host vehicle 1 in parking the target parking position.

[0081] (2) According to the parking assist method of claim 4, even if parking assist control is started after entering a place where the detection accuracy of the own position is less than the predetermined accuracy, the learning completion marker object data can be read from the storage device.

[0082] According to the parking assist method of claim 5, even if parking assist control is started after entering a place where the detection accuracy of the own position is less than a predetermined accuracy, the learning completion marker object data can be read from the storage device.

[0083] (3) According to the parking assist method described in claim 6, when the travel distance after entering a place where the detection accuracy of the own position is less than the predetermined accuracy is relatively short, the learning completion marker object data can be efficiently read from the storage device.

[0084] (4) According to the parking assist method described in claim 7, even if the driving distance after entering a place where the detection accuracy of the own position is less than the prescribed accuracy is relatively long, the learning completion mark object data can be read from the storage device.

[0085] (5) According to the parking assist method described in claim 8, it is possible to correctly select the learning completion marker object data of the target parking position registered at a location where the detection accuracy of the own position is less than the prescribed accuracy.

[0086] Explanation of symbols

[0087] 1: own vehicle, 10: parking assist device, 17: controller

Claims

1. A parking assistance method, characterized in that: Detect the vehicle's own position, estimating the detection accuracy of the own position, The self position at the point where the detection accuracy changes from a state of being greater than or equal to a predetermined accuracy to a state of being less than the predetermined accuracy is set as a first self position, detecting a relative positional relationship between a landmark object existing around the target parking position and the target parking position when the host vehicle is parked at the target parking position after the first own position is set, When the detection accuracy is less than a predetermined accuracy when the host vehicle is parked at the target parking position, the first own position is associated with the data representing the relative position relationship, i.e., the learning completion mark object data, and stored in a storage device. When the first own position is associated with the learning completion sign object data and stored, parking of the host vehicle at the target parking position is assisted based on the first own position and the learning completion sign object data.

2. The parking assistance method according to claim 1, characterized in that: detecting the own position of the host vehicle, that is, the second own position, at a time point after the learning time point at which the learning completion marker object data is stored, that is, the learning time point, When the distance between the second self-position and the first self-position becomes less than a first threshold, the learning completion marker object data stored and associated with the first self-position is read from the storage device, detecting the positions of the landmark objects existing around the host vehicle at a time point after the learning time point, that is, the positions of the surrounding landmark objects, Based on the learning completed marker object data and the positions of the surrounding marker objects, a relative position relationship between the target parking position and the current position of the host vehicle is calculated; calculating a driving trajectory from the current position of the host vehicle to the target parking position based on a relative position relationship between the target parking position and the current position of the host vehicle, Parking of the host vehicle at the target parking position is assisted based on the driving trajectory.

3. The parking assistance method according to claim 2, characterized in that: estimating the detection accuracy of the second self-position, When the detection accuracy of the second self-position changes from a state greater than the specified accuracy to a state less than the specified accuracy, and the distance between the second self-position and the first self-position becomes less than the first threshold, the learning completion mark object data associated with and stored in the first self-position is read from the storage device.

4. The parking assistance method according to claim 3, characterized in that: When the detection accuracy of the second self-position changes from a state of being greater than the prescribed accuracy to a state of being less than the prescribed accuracy, determining the first self-position existing within a range of the first threshold value from the second self-position, When an operation instructing detection of the target parking position is received from the occupant, the learning completion sign object data stored and associated with the determined first self position is read from the storage device.

5. The parking assistance method according to claim 1, characterized in that: Detecting the own position of the vehicle, i.e., the second own position, at a time point after the learning time point when the learning completion mark object data is stored, and storing the second own position when the detection accuracy of the second own position changes from a state above the specified accuracy to a state below the specified accuracy and the distance between the second own position and the first own position becomes below a first threshold, When an operation instructing to detect the target parking position is received from the occupant, the learning completion mark object data stored and associated with the first self position located within a distance range less than the first threshold value from the stored second self position is read from the storage device, detecting the positions of the landmark objects existing around the host vehicle at a time point after the learning time point, that is, the positions of the surrounding landmark objects, Based on the learning completed marker object data and the positions of the surrounding marker objects, a relative position relationship between the target parking position and the current position of the host vehicle is calculated; calculating a driving trajectory from the current position of the host vehicle to the target parking position based on a relative position relationship between the target parking position and the current position of the host vehicle, Parking of the host vehicle at the target parking position is assisted based on the driving trajectory.

6. The parking assist device according to claim 2, characterized in that: The learning completion mark object data includes the position data of the target parking position, estimating the detection accuracy of the second self-position, When the driving distance of the vehicle is below a second threshold after the detection accuracy of the second self-position changes from a state above the specified accuracy to a state below the specified accuracy, the learning completion mark object data is read from the storage device based on the self-position of the vehicle estimated by the odometry method.

7. The parking assistance method according to claim 2, characterized in that: The learned marker object data includes data of feature quantities of marker objects existing around the target parking position. estimating the detection accuracy of the second self-position, When the distance traveled by the vehicle is greater than a second threshold after the detection accuracy of the second self-position changes from a state above the prescribed accuracy to a state below the prescribed accuracy, the learning completed marker object data with feature quantities similar to the feature quantities of the surrounding marker objects in the learning completed marker object data associated with and stored in the first self-position is read from the storage device.

8. The parking assistance method according to any one of claims 2 to 7, characterized in that: storing the learning completion mark object data of a plurality of the target parking positions in the storage device, When the distance between the second self position and the first self position becomes equal to or smaller than a first threshold, the learning completion sign object data stored and associated with the first self position among the plurality of target parking positions is read from the storage device.

9. A parking assistance device, characterized in that: have: Storage device; A controller that performs the following processing: a process of acquiring position information obtained by measuring the current position of the host vehicle, i.e., the own position; a process of estimating the detection accuracy of the own position; a process of setting the own position at a location where the detection accuracy changes from a state of being greater than a prescribed accuracy to a state of being less than the prescribed accuracy as a first own position; a process of detecting a relative positional relationship between a landmark object existing around the target parking position and the target parking position when the host vehicle is parked at the target parking position after the first own position is set; When the detection accuracy is less than the specified accuracy when the vehicle is parked at the target parking position, the first self-position is associated with the data representing the relative position relationship, i.e., the learning completion mark object data, and stored in the storage device; when the first self-position is associated with the learning completion mark object data and stored, the parking of the vehicle to the target parking position is assisted based on the first self-position and the learning completion mark object data.

Images