Automatic parking method, apparatus and device, and storage medium

Abstract

An automatic parking method, apparatus and device, and a storage medium. The method comprises: acquiring a vehicle surround-view image of the current vehicle, and performing image recognition on the vehicle surround-view image, so as to obtain parking space information of a target parking space; planning a parking path on the basis of the parking space information, and controlling the current vehicle to park according to the parking path; after the parking space information meets a preset accuracy condition, updating the parking space information, wherein the preset accuracy condition is a condition in which the accuracy of the parking space information is preset to meet expectations; and performing secondary planning on the parking path on the basis of the updated parking space information, and controlling the parking of the current vehicle according to a parking path obtained after the secondary planning

Description

Automatic parking methods, devices, equipment and storage media

[0001] This application claims priority to Chinese patent application No. 202410923272.2, filed on July 10, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of intelligent driving technology, and in particular to an automatic parking method, apparatus, device and storage medium. Background Technology

[0003] With the rapid development of intelligent driving technology, automatic parking, as one of the important functions of intelligent driving systems, helps novice drivers complete parking operations safely and conveniently.

[0004] Current automatic parking systems use surround view images for parking path planning. However, since surround view images are obtained by stitching together images captured by multiple fisheye cameras after distortion correction, there is still a significant deviation between the points on the surround view image and the points in the actual world coordinate system. This may cause the vehicle to deviate from its parking position during automatic parking.

[0005] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Technical issues

[0006] The main objective of this application is to provide an automatic parking method, apparatus, device, and storage medium, which aims to solve the technical problem that existing parking methods are prone to parking deviation. Technical solutions

[0007] To achieve the above objectives, this application proposes an automatic parking method, the method comprising:

[0008] Obtain the vehicle surround view image of the current vehicle, and perform image recognition on the vehicle surround view image to obtain the parking space information of the target parking space;

[0009] The parking route is planned based on the parking space information, and the current vehicle is controlled to park according to the parking route.

[0010] After the storage location information meets the preset accuracy conditions, the storage location information is updated, wherein the preset accuracy conditions are pre-set conditions that the accuracy of the storage location information meets the desired level.

[0011] The parking path is replanned based on the updated parking space information, and the current vehicle is controlled to park according to the replanned parking path.

[0012] In one embodiment, the step of acquiring a surround view image of the current vehicle and performing image recognition on the surround view image to obtain the parking space information of the target parking space includes:

[0013] Acquire images of the surrounding environment captured by the vehicle's surround-view cameras;

[0014] The surrounding environment images are stitched together to obtain a vehicle surround view image;

[0015] The vehicle surround view image is input into a preset parking space detection model to obtain the parking space information of the target parking space.

[0016] In one embodiment, before the step of inputting the vehicle surround view image into a preset parking space detection model to obtain the parking space information of the target parking space, the method includes:

[0017] Obtain a training image dataset, wherein the training image dataset includes parking space image samples and image labels corresponding to the parking space image samples, and the image labels include parking space corner point labels and parking space separator line labels;

[0018] Based on the parking space image samples, the parking space corner labels, and the parking space separator line labels, a preset initial model is trained until the preset initial model converges to obtain a preset parking space detection model.

[0019] In one embodiment, the parking space information includes a first entrance corner point, a second entrance corner point, and a parking space dividing line recognition frame for the target parking space. The step of updating the parking space information includes:

[0020] Obtain the length of the target parking space, the first entrance coordinates of the first entrance corner point and the second entrance coordinates of the second entrance corner point, and use the direction of the parking space dividing line recognition box as the parking space direction;

[0021] The first bottom coordinates of the first bottom corner point are calculated based on the first entrance coordinates, the parking space length, and the parking space direction.

[0022] The second bottom coordinates of the second bottom corner point are calculated based on the second entrance coordinates, the parking space length, and the parking space direction.

[0023] The storage location information is updated based on the first bottom coordinates and the second bottom coordinates.

[0024] In one embodiment, the step of obtaining the length of the target parking space includes:

[0025] Obtain the parking space type of the target parking space;

[0026] Based on the parking space type, the preset parking space length mapping table is queried to obtain the parking space length of the target parking space.

[0027] In one embodiment, before the step of ensuring the storage location information meets a preset accuracy condition, the method includes:

[0028] Determine whether the first entrance corner point and the second entrance corner point are both within the field of view of the rearview camera of the current vehicle, whether the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and whether the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold.

[0029] If both the first entrance corner point and the second entrance corner point are within the field of view of the rearview camera of the current vehicle, and the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold, then the parking space information is determined to meet the preset accuracy conditions.

[0030] In one embodiment, before the step of performing secondary planning of the parking path based on the updated parking space information and controlling the current vehicle to park according to the secondary planned parking path, the method includes:

[0031] Obtain the difference in storage location orientation angle before and after the storage location information is updated;

[0032] After the angle difference of the parking space direction is greater than the preset angle difference threshold, the following steps are performed: the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path.

[0033] Once the angle difference in the parking space direction is no greater than a preset angle difference threshold, the vehicle is controlled to park according to the original parking path.

[0034] Furthermore, to achieve the above objectives, this application also proposes an automatic parking device, which includes:

[0035] The parking space identification module is used to acquire the vehicle surround view image of the current vehicle and perform image recognition on the vehicle surround view image to obtain the parking space information of the target parking space.

[0036] A primary planning module is used to plan a parking route based on the parking space information and control the current vehicle to park according to the parking route;

[0037] The storage location update module is used to update the storage location information after the storage location information meets the preset accuracy conditions, wherein the preset accuracy conditions are pre-set conditions that the accuracy of the storage location information meets the desired level.

[0038] The secondary planning module is used to perform secondary planning of the parking path based on the updated parking space information, and control the current vehicle to park according to the secondary planned parking path.

[0039] In one embodiment, the warehouse location identification module is further configured to:

[0040] Acquire images of the surrounding environment captured by the vehicle's surround-view cameras;

[0041] The surrounding environment images are stitched together to obtain a vehicle surround view image;

[0042] The vehicle surround view image is input into a preset parking space detection model to obtain the parking space information of the target parking space.

[0043] In one embodiment, the automatic parking device further includes a model building module for:

[0044] Obtain a training image dataset, wherein the training image dataset includes parking space image samples and image labels corresponding to the parking space image samples, and the image labels include parking space corner point labels and parking space separator line labels;

[0045] Based on the parking space image samples, the parking space corner labels, and the parking space separator line labels, a preset initial model is trained until the preset initial model converges to obtain a preset parking space detection model.

[0046] In one embodiment, the parking space information includes the first entrance corner point, the second entrance corner point, and the parking space dividing line recognition frame of the target parking space. The parking space update module is further used for:

[0047] Obtain the length of the target parking space, the first entrance coordinates of the first entrance corner point and the second entrance coordinates of the second entrance corner point, and use the direction of the parking space dividing line recognition box as the parking space direction;

[0048] The first bottom coordinates of the first bottom corner point are calculated based on the first entrance coordinates, the parking space length, and the parking space direction.

[0049] The second bottom coordinates of the second bottom corner point are calculated based on the second entrance coordinates, the parking space length, and the parking space direction.

[0050] The storage location information is updated based on the first bottom coordinates and the second bottom coordinates.

[0051] In one embodiment, the storage location update module is further configured to:

[0052] Obtain the parking space type of the target parking space;

[0053] Based on the parking space type, the preset parking space length mapping table is queried to obtain the parking space length of the target parking space.

[0054] In one embodiment, the storage location update module is further configured to:

[0055] Determine whether the first entrance corner point and the second entrance corner point are both within the field of view of the rearview camera of the current vehicle, whether the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and whether the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold.

[0056] If both the first entrance corner point and the second entrance corner point are within the field of view of the rearview camera of the current vehicle, and the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold, then the parking space information is determined to meet the preset accuracy conditions.

[0057] In one embodiment, the secondary planning module is further configured to:

[0058] Obtain the difference in storage location orientation angle before and after the storage location information is updated;

[0059] After the angle difference of the parking space direction is greater than the preset angle difference threshold, the following steps are performed: the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path.

[0060] Once the angle difference in the parking space direction is no greater than a preset angle difference threshold, the vehicle is controlled to park according to the original parking path.

[0061] In addition, to achieve the above objectives, this application also proposes an automatic parking device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the automatic parking method as described above.

[0062] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of the automatic parking method described above.

[0063] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the automatic parking method described above. Beneficial effects

[0064] One or more technical solutions proposed in this application have at least the following technical effects:

[0065] This application obtains the parking space information of the target parking space by acquiring a surround-view image of the current vehicle and performing image recognition on the image. Based on this parking space information, a parking path is planned, and the current vehicle is controlled to park according to the planned path. As the vehicle gets closer to the target parking space during parking, the distortion of the target parking space in the surround-view image decreases, resulting in more accurate parking space information. Therefore, this application updates the parking space information after it meets a preset accuracy condition, where the preset accuracy condition is a pre-defined condition that the accuracy of the parking space information meets expectations, thus obtaining more accurate parking space information. Then, the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path. Thus, this application effectively reduces the possibility of parking deviation during the parking process by re-planning the parking path based on the updated, more accurate parking space information after meeting the preset accuracy condition. Attached Figure Description

[0066] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0067] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 is a flowchart illustrating the automatic parking method provided in Embodiment 1 of this application;

[0069] Figure 2 is a schematic diagram of the target parking space involved in the embodiments of this application;

[0070] Figure 3 is a flowchart of the automatic parking method provided in Embodiment 2 of this application;

[0071] Figure 4 is a schematic diagram of a parking scenario involved in an embodiment of this application;

[0072] Figure 5 is a flowchart of the automatic parking method provided in Embodiment 3 of this application;

[0073] Figure 6 is a schematic diagram of the module structure of the automatic parking device according to an embodiment of this application;

[0074] Figure 7 is a schematic diagram of the hardware operating environment involved in the automatic parking method in this application embodiment.

[0075] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Embodiments of the present invention

[0076] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0077] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0078] The main solution of this application embodiment is as follows: acquire a vehicle surround view image of the current vehicle, and perform image recognition on the vehicle surround view image to obtain the parking space information of the target parking space; plan a parking path according to the parking space information, and control the current vehicle to park according to the parking path; after the parking space information meets a preset accuracy condition, update the parking space information, wherein the preset accuracy condition is a pre-set condition that the accuracy of the parking space information meets the desired condition; perform secondary planning on the parking path according to the updated parking space information, and control the current vehicle to park according to the secondary planned parking path.

[0079] In this embodiment, for ease of description, the vehicle controller will be used as the execution subject in the following description.

[0080] This application provides a solution whereby, as the vehicle gets closer to the target parking space during parking, the distortion of the target parking space in the vehicle's surround-view image decreases, and the parking space information becomes more accurate. Therefore, this application updates the parking space information after it meets a preset accuracy condition, where the preset accuracy condition is a pre-defined condition that the accuracy of the parking space information meets expectations. Then, based on the updated parking space information, a secondary parking path is planned, and the vehicle is controlled to park according to the secondary planned parking path. Thus, by performing secondary parking path planning based on the updated, more accurate parking space information after meeting the preset accuracy condition, this application effectively reduces the possibility of parking deviation during the parking process.

[0081] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device, vehicle controller, or vehicle computer capable of performing the above functions. The following description uses a vehicle controller as an example to illustrate this embodiment and the subsequent embodiments.

[0082] Based on this, this application provides an automatic parking method. Referring to FIG1, FIG1 is a flowchart of the first embodiment of the automatic parking method of this application.

[0083] In this embodiment, the automatic parking method includes steps S10 to S40:

[0084] Step S10: Obtain the vehicle surround view image of the current vehicle, and perform image recognition on the vehicle surround view image to obtain the parking space information of the target parking space.

[0085] It should be noted that the vehicle surround view image can be a BEV (Bird's Eye View) image, which can provide a complete view of the current vehicle's surrounding environment, including the front, rear, sides, and top of the vehicle.

[0086] Additionally, it should be noted that, referring to Figure 2, which is a schematic diagram of the target parking space scenario involved in the embodiments of this application, the parking space information may include the four corner points of the target parking space (i.e., points 1, 2, 3, and 4 in Figure 2) and two parking space separation lines.

[0087] This embodiment acquires surrounding environment images captured by the front-view camera, left-side camera, right-side camera, and rear-view camera of the current vehicle, and then stitches these images together according to their positions to obtain a vehicle surround view image. A predetermined parking space detection algorithm is then used to perform image recognition on the vehicle surround view image to obtain the parking space information of the target parking space. For example, the parking space detection algorithm can be a feature-based detection method, a machine learning-based detection method (such as a classifier or regressor), or a deep learning-based detection method (such as convolutional neural networks, YOLO (You Only Look Once) model, SSD (Single Shot MultiBox Detector) model, etc.).

[0088] In one feasible implementation, step S10 may include steps S11 to S13:

[0089] Step S11: Obtain images of the surrounding environment captured by the surround-view camera of the current vehicle;

[0090] Step S12: The surrounding environment images are stitched together to obtain a vehicle surround view image;

[0091] Step S13: Input the vehicle surround view image into the preset parking space detection model to obtain the parking space information of the target parking space.

[0092] It should be noted that the surround-view camera may include a front-view camera for capturing images of the front of the vehicle, a left-side camera for capturing images of the left side of the vehicle, a right-side camera for capturing images of the right side of the vehicle, and a rear-view camera for capturing images of the rear of the vehicle. It is understandable that, in order to obtain a wider field of view, the front-view camera, left-side camera, right-side camera, and rear-view camera are generally fisheye cameras.

[0093] It should also be noted that the preset parking space detection model is a pre-trained neural network model for detecting parking spaces. This preset parking space detection model can be a convolutional neural network model, a YOLO model, an SSD model, etc. For example, since parking space isolation lines may be rotated or tilted in the image, a conventional horizontal rectangular bounding box may not tightly enclose the target (i.e., the parking space isolation line), thus affecting the accuracy and precision of the detection. Therefore, the preset parking space detection model can adopt the YOLOv5_OBB (Oriented Bounding Box) model, which can adjust according to the rotation angle of the target, thereby more tightly enclosing the target and improving the accuracy and precision of the detection.

[0094] This embodiment acquires surrounding environment images captured by the front-view camera, left-side camera, right-side camera, and rear-view camera of the current vehicle. These images are then stitched together according to their positions to obtain a vehicle surround view image. Furthermore, since fisheye cameras exhibit significant image distortion, distortion correction processing can be performed before stitching to reduce the distortion in the vehicle surround view image and improve the accuracy of the identified parking space information. This distortion correction processing can be based on methods such as filtering based on local image features, convolutional kernel correction, or correction based on predetermined distortion parameters; this embodiment does not limit the specific methods used. The vehicle surround view image can then be input into a preset parking space detection model to obtain the parking space information of the target parking space.

[0095] In one feasible implementation, steps A10 to A20 may be included before step S13:

[0096] Step A10: Obtain a training image dataset, wherein the training image dataset includes parking space image samples and image labels corresponding to the parking space image samples, and the image labels include parking space corner point labels and parking space dividing line labels;

[0097] Step A20: Based on the parking space image samples, the parking space corner labels, and the parking space separator line labels, train the preset initial model until the preset initial model converges to obtain the preset parking space detection model.

[0098] It should be noted that the training image dataset includes parking space image samples and corresponding image labels. The image labels include parking space corner point labels and parking space separator line labels. The parking space image samples are panoramic images including parking spaces. The parking space corner point labels are the marked corner points of the parking spaces in the parking space image samples, and the parking space separator line labels are the marked parking space separator line recognition boxes in the parking space image samples. The preset initial model is a pre-initialized neural network model, such as a convolutional neural network, YOLO model, or SSD model.

[0099] In this embodiment, a predetermined number of surround view images (i.e., images from the perspective of a BEV vehicle) are acquired, and the corner points and dividing lines of parking spaces in the surround view images are labeled. Corner points are represented as points, and dividing lines as bounding boxes, thereby generating corner point labels and dividing line labels to obtain a training image dataset. The parking space image samples are then input into a preset initial model to obtain training detection results. A loss function value is then calculated based on the training detection results and the corner point and dividing line labels. The model parameters of the preset initial model are updated and iterated based on the loss function value until the preset initial model converges (i.e., the number of iterations reaches a preset maximum value, or the loss function value is minimized). This preset initial model is then used as the preset parking space detection model. In this embodiment, a detection model capable of accurately identifying corner points and dividing line bounding boxes in parking spaces is obtained by training parking space image samples and their corresponding corner point and dividing line labels.

[0100] Step S20: Plan a parking route based on the parking space information, and control the current vehicle to park according to the parking route.

[0101] In this embodiment, a parking path can be planned based on the parking space information. For example, a parking path that satisfies the parking space information is generated based on the parking space information, environmental information, vehicle information, and parking specifications. For example, the parking specifications are pre-set constraints on the parking path, which may include: conforming to the laws of vehicle structural motion and not exceeding the physical requirements of vehicle dynamics or kinematics; preventing collisions or overshooting of parking spaces during parking; ensuring the parking path is as simple as possible; and ensuring a smooth parking process. After generating the parking path, the current vehicle can be controlled to park according to the parking path.

[0102] Step S30: After the storage location information meets the preset accuracy conditions, update the storage location information, wherein the preset accuracy conditions are pre-set conditions that the accuracy of the storage location information meets the desired level.

[0103] It should be noted that the preset accuracy condition is a pre-set condition that the accuracy of the parking space information meets the desired level. When the parking space information meets the preset accuracy condition, the current vehicle is in a specified posture and position relative to the target parking space, so that the parking space information identified by the current vehicle in the specified posture and position meets the accuracy requirements for the parking space information. For example, the specified posture and position may be that the angle between the parking space direction of the target parking space and the longitudinal axis direction of the vehicle body is less than a predetermined angle threshold (such as 5°, 10°, 15°, etc.), and the distance between the entrance end (corner point or edge on the entrance side) of the target parking space and the current vehicle is within a predetermined distance range (such as 0.3~0.8m). For example, to further improve the accuracy of the identified parking space information, the preset accuracy conditions may include: both the first and second entrance corners of the target parking space are within the field of view of the rearview camera of the current vehicle; the length of the long side of the parking space divider recognition frame of the target parking space is greater than a preset length threshold; the included angle between the long sides of the parking space divider recognition frame is less than a preset included angle threshold; and the distance between the rear edge line of the current vehicle and the line connecting the first and second entrance corners is less than a preset distance threshold.

[0104] This embodiment determines whether the first and second entrance corner points are both within the field of view of the rearview camera of the current vehicle, whether the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and whether the angle between the long sides of the parking space dividing line recognition frame is less than a preset angle threshold. If the first and second entrance corner points are both within the field of view of the rearview camera of the current vehicle, and the length of the long side of the parking space dividing line recognition frame is greater than the preset length threshold, and the angle between the long sides of the parking space dividing line recognition frame is less than the preset angle threshold, then the parking space information is determined to meet the preset accuracy conditions. After the parking space information meets the preset accuracy conditions, the parking space information is updated. The way to update the parking space information can be to use the four parking space corner points and two parking space dividing lines of the target parking space after meeting the preset accuracy conditions as the new parking space information. Alternatively, the first bottom coordinates of the first bottom corner point can be calculated based on the first entrance coordinates, the parking space length, and the parking space direction; and the second bottom coordinates of the second bottom corner point can be calculated based on the second entrance coordinates, the parking space length, and the parking space direction. The first and second bottom coordinates calculated in this way are more accurate, and the storage location information is then updated based on the first and second bottom coordinates.

[0105] Step S40: Perform secondary planning of the parking path based on the updated parking space information, and control the current vehicle to park according to the secondary planned parking path.

[0106] This embodiment can perform secondary planning of the parking path based on the updated parking space information. For example, based on the updated parking space information, environmental information, vehicle information, and parking specifications, a parking path that meets the updated parking space information is generated. Then, the vehicle is controlled to park according to the secondary planned parking path.

[0107] The first embodiment of this application provides an automatic parking method. It obtains the parking space information of a target parking space by acquiring a surround-view image of the current vehicle and performing image recognition on the image. A parking path is planned based on the parking space information, and the current vehicle is controlled to park according to the parking path. As the vehicle gets closer to the target parking space during parking, the distortion of the target parking space in the surround-view image decreases, and the parking space information becomes more accurate. Therefore, in this embodiment, after the parking space information meets a preset accuracy condition, the parking space information is updated. This preset accuracy condition is a pre-set condition that the accuracy of the parking space information meets a desired level to obtain more accurate parking space information. Then, the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path. Thus, this embodiment effectively reduces the possibility of parking deviation during the parking process by re-planning the parking path based on the updated, more accurate parking space information after meeting the preset accuracy condition.

[0108] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to Figure 3. The parking space information includes the first entrance corner point, the second entrance corner point, and the parking space dividing line identification frame of the target parking space. The step of updating the parking space information in step S30 includes steps S31 to S34:

[0109] Step S31: Obtain the length of the target parking space, the first entrance coordinates of the first entrance corner point and the second entrance coordinates of the second entrance corner point, and use the direction of the parking space dividing line recognition box as the parking space direction;

[0110] Step S32: Calculate the first bottom coordinates of the first bottom corner point based on the first entrance coordinates, the parking space length, and the parking space direction;

[0111] Step S33: Calculate the second bottom coordinates of the second bottom corner point based on the second entrance coordinates, the parking space length, and the parking space direction;

[0112] Step S34: Update the storage location information based on the first bottom coordinates and the second bottom coordinates.

[0113] It should be noted that the parking space length refers to the length extending from the first or second entrance corner point into the parking space to the bottom of the space. Since parking space design has specific requirements, the parking space length can be a pre-set value. The first entrance corner point is one corner point at the end where the vehicle enters the target parking space, and the second entrance corner point is another corner point at the same end. The parking space direction refers to the orientation of the parking space dividing line recognition frame in the coordinate system (generally the world coordinate system) corresponding to the coordinates of the first and second entrances.

[0114] Referring to Figure 4, which is a schematic diagram of a parking scenario according to an embodiment of this application, this embodiment obtains the length of the target parking space, the first entrance coordinates of the first entrance corner point (i.e., parking space corner point 1 in Figure 4), and the second entrance coordinates of the second entrance corner point (i.e., parking space corner point 2 in Figure 4), and uses the direction of the parking space dividing line recognition frame (i.e., the long side of the parking space dividing line recognition frame or the direction of the corresponding parking space dividing line) as the parking space direction. Then, based on the first entrance coordinates, the parking space length, and the parking space direction, the first bottom coordinates of the first bottom corner point (i.e., parking space corner point 4 in Figure 4) are calculated, and based on the second entrance coordinates, the parking space length, and the parking space direction, the second bottom coordinates of the second bottom corner point (i.e., parking space corner point 3 in Figure 4) are calculated. Taking the first entrance corner point as (x1, y1), the second entrance corner point as (x2, y2), the first bottom corner point as (x3, y3), and the second bottom corner point as (x4, y4) as an example, the calculation methods for the first and second bottom coordinates are as follows:

[0115] x3 = x2 + Lcos(θ);

[0116] y3 = y2 + Lsin(θ);

[0117] x4 = x1 + Lcos(θ);

[0118] y4 = y1 + Lsin(θ).

[0119] Where L is the length of the parking space, and θ is the angle of the parking space direction in the coordinate system (generally the world coordinate system) corresponding to the coordinates of the first entrance and the second entrance.

[0120] Because the first and second entrance corner points are closer to the rearview camera of the current vehicle, the image distortion is smaller, resulting in higher accuracy of the obtained first and second entrance coordinates. Therefore, the first and second bottom coordinates, calculated directly from the first and second entrance coordinates, along with the parking space length and orientation, are more accurate than those obtained through image mapping. Furthermore, the parking space information is updated based on the first and second bottom coordinates, replacing the coordinates of the first and second bottom corner points in the parking space information. It is understood that the parking space information is updated according to changes in the vehicle's surround-view image.

[0121] In one feasible implementation, step S31 includes steps B10 to B20:

[0122] Step B10: Obtain the parking space type of the target parking space;

[0123] Step B20: Based on the parking space type, query the preset parking space length mapping table to obtain the parking space length of the target parking space.

[0124] It should be noted that the parking space types may include parallel parking spaces (the parking space direction is parallel to the road direction), perpendicular parking spaces (the parking space direction is perpendicular to the road direction), and angled parking spaces (the parking space direction is at a certain angle to the road direction). The preset parking space length mapping table is used to describe the mapping relationship between parking space types and parking space lengths. For example, the lengths of perpendicular and angled parking spaces are between 5 and 6 meters (e.g., 5.5 meters), while the lengths of parallel parking spaces are between 2.2 and 2.5 meters (e.g., 2.5 meters) because one enters the parking space from its long side.

[0125] This embodiment can obtain the parking space type of the target parking space; based on the parking space type, it queries a preset parking space length mapping table to obtain the length of the target parking space. Then, based on the first entrance coordinates, the second entrance coordinates, the parking space length, and the parking space direction, it calculates the first bottom coordinates of the first bottom corner point and the second bottom coordinates of the second bottom corner point. This embodiment can avoid deviations in the calculation of the coordinates of the first and second bottom corner points due to different parking space types, thus preventing parking deviations.

[0126] In one feasible implementation, steps C10 to C20 are included before step S30:

[0127] Step C10: Determine whether the first entrance corner point and the second entrance corner point are both within the field of view of the rearview camera of the current vehicle, whether the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and whether the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold.

[0128] Step C20: If both the first entrance corner point and the second entrance corner point are within the field of view of the rearview camera of the current vehicle, and the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold, then the parking space information is determined to meet the preset accuracy conditions.

[0129] This embodiment uses the preset accuracy conditions to ensure the accuracy of the coordinates of the first and second entrance corner points and the direction of the storage location in the identified storage location information, so as to calculate more accurate coordinates of the first and second bottom corner points.

[0130] Since the target parking space has minimal distortion within the field of view of the rearview camera, this embodiment determines whether both the first and second entrance corner points are within the field of view of the rearview camera of the current vehicle; and determines whether the length of the long side of the parking space dividing line recognition frame (i.e., the length of the longest side of the parking space dividing line recognition frame) is greater than a preset length threshold (e.g., 0.5m, 0.6m, 0.8m, etc.) to avoid recognition deviations in the direction of the parking space dividing line recognition frame (i.e., the parking space direction) with shorter long side lengths; since the long sides of the parking space dividing line recognition frame are parallel in actual scenarios, it also determines whether the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold (e.g., 2°, 3°, etc.) to determine whether the long sides of the parking space dividing line recognition frame are parallel, and to determine whether there is significant distortion in the parking space dividing line recognition frame. If both the first and second entrance corner points are within the field of view of the rearview camera of the current vehicle, it indicates that the coordinates of the first and second entrance corner points are highly accurate. If the length of the long side of the parking space divider recognition frame is greater than a preset length threshold, it avoids recognition deviations in the direction of the parking space divider recognition frame (i.e., the parking space direction) when the long side is shorter. If the angle between the long sides of the parking space divider recognition frame is less than a preset angle threshold, it indicates that the parking space divider recognition frame does not have significant distortion, thus determining that the parking space information meets the preset accuracy conditions. If the first and second entrance corner points are not both within the field of view of the rearview camera of the current vehicle, and / or the length of the long side of the parking space divider recognition frame is not greater than the preset length threshold, and / or the angle between the long sides of the parking space divider recognition frame is not less than the preset angle threshold, it is determined that the parking space information does not meet the preset accuracy conditions, and the current vehicle is controlled to park according to the original parking path.

[0131] The second embodiment of this application provides an automatic parking method. In this embodiment, after the parking space information meets preset accuracy conditions, the length of the target parking space, the first entrance coordinates of the first entrance corner point, and the second entrance coordinates of the second entrance corner point are obtained, and the direction of the parking space divider recognition frame is used as the parking space direction. Based on the first entrance coordinates, the parking space length, and the parking space direction, the first bottom coordinates of the first bottom corner point are calculated; based on the second entrance coordinates, the parking space length, and the parking space direction, the second bottom coordinates of the second bottom corner point are calculated; and based on the first bottom coordinates and the second bottom coordinates, the parking space information is updated. Therefore, this embodiment, after obtaining the first entrance corner point, the second entrance corner point, and the parking space angle with the required accuracy, directly calculates the first and second bottom corner points using the first entrance corner point, the second entrance corner point, the parking space length, and the parking space direction, which is more accurate than obtaining the coordinates of the first and second bottom corner points through image mapping. Furthermore, based on the first bottom coordinates and the second bottom coordinates, the parking space information is updated, resulting in new parking space information that is more accurate than the previous parking space information.

[0132] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description and will not be repeated hereafter. Based on this, please refer to Figure 5; steps D10 to D30 are included before step S40:

[0133] Step D10: Obtain the difference in storage location direction angle before and after the storage location information is updated;

[0134] Step D20: After the angle difference of the parking space direction is greater than the preset angle difference threshold, the following steps are performed: the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path.

[0135] Step D30: After the angle difference in the parking space direction is not greater than the preset angle difference threshold, control the current vehicle to park according to the original parking path.

[0136] It should be noted that the storage location direction angle difference is the angle difference between the storage location direction in the storage location information when planning the parking path and the storage location direction in the updated storage location information.

[0137] This embodiment addresses the issue of excessive parking space re-planning after a second planning of the parking path. It calculates the angle difference between the parking space directions in the planned parking space information and the updated parking space information. Then, it determines whether this angle difference is greater than a preset angle difference threshold (e.g., 0.3, 0.5, 0.8, etc.). If the angle difference is greater than the preset threshold, it indicates a significant deviation in the parking space direction before and after the update. In this case, the following steps are performed: the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path. If the angle difference is not greater than the preset threshold, it indicates a small deviation in the parking space direction before and after the update. In this case, the current vehicle is controlled to park according to the original parking path.

[0138] The third embodiment of this application provides an automatic parking method. This method involves obtaining the difference in parking space orientation angles before and after the parking space information is updated. If the difference in orientation angles exceeds a preset threshold, the following steps are performed: a secondary parking path is planned based on the updated parking space information, and the current vehicle is controlled to park according to the secondary planned parking path. If the difference in orientation angles is not greater than the preset threshold, the current vehicle is controlled to park according to the original parking path. Therefore, this embodiment maintains the original parking path even when the deviation in parking space orientations before and after the update is small, reducing unnecessary parking maneuvers and effectively improving parking efficiency.

[0139] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the automatic parking method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0140] This application also provides an automatic parking device, as shown in Figure 6, the automatic parking device comprising:

[0141] The parking space identification module 10 is used to acquire the vehicle surround view image of the current vehicle and perform image recognition on the vehicle surround view image to obtain the parking space information of the target parking space.

[0142] The primary planning module 20 is used to plan a parking path based on the parking space information and control the current vehicle to park according to the parking path;

[0143] The storage location update module 30 is used to update the storage location information after the storage location information meets the preset accuracy conditions, wherein the preset accuracy conditions are pre-set conditions that the accuracy of the storage location information meets the expected conditions.

[0144] The secondary planning module 40 is used to perform secondary planning on the parking path based on the updated parking space information, and control the current vehicle to park according to the secondary planned parking path.

[0145] In one embodiment, the warehouse location identification module 10 is further configured to:

[0146] Acquire images of the surrounding environment captured by the vehicle's surround-view cameras;

[0147] The surrounding environment images are stitched together to obtain a vehicle surround view image;

[0148] The vehicle surround view image is input into a preset parking space detection model to obtain the parking space information of the target parking space.

[0149] In one embodiment, the automatic parking device further includes a model building module for:

[0150] Obtain a training image dataset, wherein the training image dataset includes parking space image samples and image labels corresponding to the parking space image samples, and the image labels include parking space corner point labels and parking space separator line labels;

[0151] Based on the parking space image samples, the parking space corner labels, and the parking space separator line labels, a preset initial model is trained until the preset initial model converges to obtain a preset parking space detection model.

[0152] In one embodiment, the parking space information includes the first entrance corner point, the second entrance corner point, and the parking space dividing line recognition frame of the target parking space. The parking space update module 30 is further used for:

[0153] Obtain the length of the target parking space, the first entrance coordinates of the first entrance corner point and the second entrance coordinates of the second entrance corner point, and use the direction of the parking space dividing line recognition box as the parking space direction;

[0154] The first bottom coordinates of the first bottom corner point are calculated based on the first entrance coordinates, the parking space length, and the parking space direction.

[0155] The second bottom coordinates of the second bottom corner point are calculated based on the second entrance coordinates, the parking space length, and the parking space direction.

[0156] The storage location information is updated based on the first bottom coordinates and the second bottom coordinates.

[0157] In one embodiment, the storage location update module 30 is further configured to:

[0158] Obtain the parking space type of the target parking space;

[0159] Based on the parking space type, the preset parking space length mapping table is queried to obtain the parking space length of the target parking space.

[0160] In one embodiment, the storage location update module 30 is further configured to:

[0161] Determine whether the first entrance corner point and the second entrance corner point are both within the field of view of the rearview camera of the current vehicle, whether the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and whether the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold.

[0162] If both the first entrance corner point and the second entrance corner point are within the field of view of the rearview camera of the current vehicle, and the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold, then the parking space information is determined to meet the preset accuracy conditions.

[0163] In one embodiment, the secondary planning module 40 is further configured to:

[0164] Obtain the difference in storage location orientation angle before and after the storage location information is updated;

[0165] After the angle difference of the parking space direction is greater than the preset angle difference threshold, the following steps are performed: the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path.

[0166] Once the angle difference in the parking space direction is no greater than a preset angle difference threshold, the vehicle is controlled to park according to the original parking path.

[0167] The device provided in this application, employing the automatic parking method in the above embodiments, can solve the technical problem of parking deviation that easily occurs in existing parking methods. Compared with the prior art, the beneficial effects of the device provided in this application are the same as those of the automatic parking method provided in the above embodiments, and other technical features in the automatic parking device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0168] This application provides an automatic parking device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the automatic parking method in Embodiment 1 above.

[0169] Referring now to Figure 7, a structural schematic diagram of an automatic parking device suitable for implementing embodiments of this application is shown. The automatic parking device in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The automatic parking device shown in Figure 7 is merely an example and should not impose any limitations on the functionality and scope of use of embodiments of this application.

[0170] As shown in Figure 7, the automatic parking device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the automatic parking device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the automatic parking equipment to communicate wirelessly or wiredly with other devices to exchange data. Although the figures show automatic parking equipment with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented alternatively.

[0171] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0172] The automatic parking device provided in this application, employing the automatic parking method described in the above embodiments, can solve the technical problem of parking deviation that easily occurs in existing parking methods. Compared with the prior art, the beneficial effects of the automatic parking device provided in this application are the same as those of the automatic parking method provided in the above embodiments, and other technical features of this automatic parking device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0173] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0174] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0175] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the automatic parking method in the above embodiments.

[0176] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or apparatus. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0177] The aforementioned computer-readable storage medium may be included in the automatic parking device; or it may exist independently and not be installed in the automatic parking device.

[0178] The aforementioned computer-readable storage medium carries one or more programs that, when executed by the automatic parking device, cause the automatic parking device to: a parking space identification module for acquiring a surround-view image of the current vehicle and performing image recognition on the surround-view image to obtain parking space information of the target parking space; a primary planning module for planning a parking path based on the parking space information and controlling the current vehicle to park according to the parking path; a parking space update module for updating the parking space information after the parking space information meets a preset accuracy condition, wherein the preset accuracy condition is a pre-set condition that the accuracy of the parking space information meets expectations; and a secondary planning module for performing secondary planning on the parking path based on the updated parking space information and controlling the current vehicle to park according to the secondary planned parking path.

[0179] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0180] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0181] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0182] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described automatic parking method, which can solve the technical problem of parking deviation that easily occurs in existing parking methods. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the automatic parking method provided in the above embodiments, and will not be repeated here.

[0183] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the automatic parking method described above.

[0184] The computer program product provided in this application can solve the technical problem of parking deviation that easily occurs in existing parking methods. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the automatic parking method provided in the above embodiments, and will not be repeated here.

[0185] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. An automatic parking method, wherein, The method includes: Obtain the vehicle surround view image of the current vehicle, and perform image recognition on the vehicle surround view image to obtain the parking space information of the target parking space; The parking route is planned based on the parking space information, and the current vehicle is controlled to park according to the parking route. After the storage location information meets the preset accuracy conditions, the storage location information is updated, wherein the preset accuracy conditions are pre-set conditions that the accuracy of the storage location information meets the desired level. The parking path is replanned based on the updated parking space information, and the current vehicle is controlled to park according to the replanned parking path.

2. The method as described in claim 1, wherein, The steps of obtaining the vehicle surround view image of the current vehicle and performing image recognition on the vehicle surround view image to obtain the parking space information of the target parking space include: Acquire images of the surrounding environment captured by the vehicle's surround-view cameras; The surrounding environment images are stitched together to obtain a vehicle surround view image; The vehicle surround view image is input into a preset parking space detection model to obtain the parking space information of the target parking space.

3. The method as described in claim 2, wherein, Before the step of inputting the vehicle surround view image into a preset parking space detection model to obtain the parking space information of the target parking space, the method includes: Obtain a training image dataset, wherein the training image dataset includes parking space image samples and image labels corresponding to the parking space image samples, and the image labels include parking space corner point labels and parking space separator line labels; Based on the parking space image samples, the parking space corner labels, and the parking space separator line labels, a preset initial model is trained until the preset initial model converges to obtain a preset parking space detection model.

4. The method of claim 1, wherein, The parking space information includes the first entrance corner point, the second entrance corner point, and the parking space dividing line recognition frame of the target parking space. The step of updating the parking space information includes: Obtain the length of the target parking space, the first entrance coordinates of the first entrance corner point and the second entrance coordinates of the second entrance corner point, and use the direction of the parking space dividing line recognition box as the parking space direction; The first bottom coordinates of the first bottom corner point are calculated based on the first entrance coordinates, the parking space length, and the parking space direction. The second bottom coordinates of the second bottom corner point are calculated based on the second entrance coordinates, the parking space length, and the parking space direction. The storage location information is updated based on the first bottom coordinates and the second bottom coordinates.

5. The method of claim 4, wherein, The step of obtaining the length of the target parking space includes: Obtain the parking space type of the target parking space; Based on the parking space type, the preset parking space length mapping table is queried to obtain the parking space length of the target parking space.

6. The method of claim 4, wherein, Before the step where the storage location information meets the preset accuracy conditions, the method includes: Determine whether the first entrance corner point and the second entrance corner point are both within the field of view of the rearview camera of the current vehicle, whether the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and whether the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold. If both the first entrance corner point and the second entrance corner point are within the field of view of the rearview camera of the current vehicle, and the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold, then the parking space information is determined to meet the preset accuracy conditions.

7. The method according to any one of claims 1 to 6, wherein, Before the step of performing secondary planning of the parking path based on the updated parking space information and controlling the current vehicle to park according to the secondary planned parking path, the method includes: Obtain the difference in storage location orientation angle before and after the storage location information is updated; After the angle difference of the parking space direction is greater than the preset angle difference threshold, the following steps are performed: the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path. Once the angle difference in the parking space direction is no greater than a preset angle difference threshold, the vehicle is controlled to park according to the original parking path.

8. An automatic parking device, wherein, The automatic parking device includes: The parking space identification module is used to acquire the vehicle surround view image of the current vehicle and perform image recognition on the vehicle surround view image to obtain the parking space information of the target parking space. A primary planning module is used to plan a parking route based on the parking space information and control the current vehicle to park according to the parking route; The storage location update module is used to update the storage location information after the storage location information meets the preset accuracy conditions, wherein the preset accuracy conditions are pre-set conditions that the accuracy of the storage location information meets the desired level. The secondary planning module is used to perform secondary planning of the parking path based on the updated parking space information, and control the current vehicle to park according to the secondary planned parking path.

9. The apparatus of claim 8, wherein, The warehouse location identification module is also used for: Acquire images of the surrounding environment captured by the vehicle's surround-view cameras; The surrounding environment images are stitched together to obtain a vehicle surround view image; The vehicle surround view image is input into a preset parking space detection model to obtain the parking space information of the target parking space.

10. The apparatus of claim 9, wherein, The automatic parking device also includes a model building module for: Obtain a training image dataset, wherein the training image dataset includes parking space image samples and image labels corresponding to the parking space image samples, and the image labels include parking space corner point labels and parking space separator line labels; Based on the parking space image samples, the parking space corner labels, and the parking space separator line labels, a preset initial model is trained until the preset initial model converges to obtain a preset parking space detection model.

11. The apparatus of claim 8, wherein, The parking space information includes the first entrance corner point, the second entrance corner point, and the parking space dividing line recognition frame of the target parking space. The parking space update module is also used for: Obtain the length of the target parking space, the first entrance coordinates of the first entrance corner point and the second entrance coordinates of the second entrance corner point, and use the direction of the parking space dividing line recognition box as the parking space direction; The first bottom coordinates of the first bottom corner point are calculated based on the first entrance coordinates, the parking space length, and the parking space direction. The second bottom coordinates of the second bottom corner point are calculated based on the second entrance coordinates, the parking space length, and the parking space direction. The storage location information is updated based on the first bottom coordinates and the second bottom coordinates.

12. The apparatus of claim 11, wherein, The storage location update module is also used for: Obtain the parking space type of the target parking space; Based on the parking space type, the preset parking space length mapping table is queried to obtain the parking space length of the target parking space.

13. The apparatus of claim 11, wherein, The storage location update module is also used for: Determine whether the first entrance corner point and the second entrance corner point are both within the field of view of the rearview camera of the current vehicle, whether the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and whether the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold. If both the first entrance corner point and the second entrance corner point are within the field of view of the rearview camera of the current vehicle, and the length of the long side of the parking space dividing line recognition frame is greater than a preset length threshold, and the included angle between the long sides of the parking space dividing line recognition frame is less than a preset included angle threshold, then the parking space information is determined to meet the preset accuracy conditions.

14. The apparatus according to any one of claims 8 to 13, wherein, The quadratic programming module is also used for: Obtain the difference in storage location orientation angle before and after the storage location information is updated; After the angle difference of the parking space direction is greater than the preset angle difference threshold, the following steps are performed: the parking path is re-planned based on the updated parking space information, and the current vehicle is controlled to park according to the re-planned parking path. Once the angle difference in the parking space direction is no greater than a preset angle difference threshold, the vehicle is controlled to park according to the original parking path.

15. An automatic parking device, wherein, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the automatic parking method as described in any one of claims 1 to 7.

16. A storage medium, wherein, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the automatic parking method as described in any one of claims 1 to 7.

Images