Charging port searching method and device, computer equipment and storage medium

Abstract

The embodiment of the invention discloses a charging port searching method and device, computer equipment and a storage medium. The method is used for searching a target charging port of a target charging vehicle in a target parking space, a charging cover of the target charging vehicle is in an open state, and the method comprises the steps of obtaining a first point cloud image of a region corresponding to the target parking space through a first camera; through a field potential method and the first point cloud image, identification processing of a charging port area is carried out on the target charging vehicle, the position of the charging port area is obtained, and the charging port area comprises a charging cover and a target charging port; moving the charging mechanical arm to an image acquisition area corresponding to the position of the charging port area; when the charging mechanical arm is located in the image collection area, a second point cloud image is obtained through a second camera in the charging mechanical arm; and performing charging port position identification processing on the target charging port according to a preset charging port standard model and the second point cloud image to obtain a target charging port position. According to the scheme, the charging port of the electric vehicle can be automatically searched.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a charging port searching method and device, a computer device and a storage medium. BACKGROUND

[0002] With the increase in the number of electric vehicles, the rapid development of electric vehicles provides new opportunities for the charging facility industry. When a traditional electric vehicle is charging, the docking process of the charging port needs to be completed manually, and improper manual operation may damage the charging port. Manual operation has potential electric shock safety hazards.

[0003] To solve the above problems, an unmanned automatic charging technology for automatically plugging and unplugging a charging gun is needed, and accurately identifying a charging port is a key technology for unmanned automatic charging. In order to realize the unmanned automatic charging technology for automatically plugging and unplugging a charging gun, a method for automatically searching for a charging port of an electric vehicle is urgently needed. SUMMARY

[0004] The present application provides a charging port searching method and device, a computer device and a storage medium, which can automatically search for a charging port of an electric vehicle.

[0005] In a first aspect, the present application provides a charging port searching method, which is used to search for a target charging port of a target charging vehicle. The target charging vehicle is a to-be-charged vehicle parked in a target parking space, and a charging cover of the target charging vehicle is in an open state. The method comprises the following steps: obtaining a first point cloud image of a region corresponding to the target parking space through a first camera; performing a charging port region identification process on the target charging vehicle through a preset field potential method and the first point cloud image to obtain a charging port region position corresponding to the charging port region, wherein the charging port region comprises the charging cover and the target charging port; moving a charging mechanical arm to an image acquisition region corresponding to the charging port region position; obtaining a second point cloud image through a second camera in the charging mechanical arm when the charging mechanical arm is in the image acquisition region; performing a charging port position identification process on the target charging port according to a preset charging port standard model and the second point cloud image to obtain a target charging port position.

[0006] In some embodiments, the step of performing a charging port region identification process on the target charging vehicle through a preset field potential method and the first point cloud image to obtain a charging port region position corresponding to the charging port region comprises: clip the first point cloud image according to a depth dimension image in the first point cloud image, to obtain a clipped first point cloud image; traverse point clouds in a horizontal dimension image and a vertical dimension image in the clipped first point cloud image, to extract a plurality of first transition points; filter non-target points in the plurality of first transition points according to a preset first transition threshold interval, and obtain a plurality of candidate transition point sets through clustering; determine discrete centers of the plurality of candidate transition point sets; respectively calculate distance values between transition centers of each candidate transition point set and the discrete centers; determine a target transition point set from the plurality of candidate transition point sets according to a preset distance threshold and the distance values respectively corresponding to each candidate transition point set, and determine a position of the target transition point set as the charging port region position.

[0007] In some embodiments, the determining of the target transition point set from the plurality of candidate transition point sets according to the preset distance threshold and the distance values respectively corresponding to each candidate transition point set comprises: filtering candidate transition point sets with distance values greater than the distance threshold from the plurality of candidate transition point sets, to obtain filtered candidate transition point sets; if there are a plurality of filtered candidate transition point sets, determining a filtered candidate transition point set with the most transition points from the plurality of filtered candidate transition point sets as the target transition point set; if there is only one filtered candidate transition point set, determining the filtered candidate transition point set as the target transition point set.

[0008] In some embodiments, the charging port position recognition processing of the target charging port according to the preset charging port standard model and the second point cloud image comprises: determining a charging port center position in a depth dimension image of the second point cloud image according to a scalable 2D model of the charging port standard model and a depth dimension image in the second point cloud image; constructing a charging port target search region in the depth dimension image of the second point cloud image with the charging port center position as the center and with a preset region size; extracting a charging port contour from the charging port target search region according to a preset charging port depth threshold; mapping the charging port contour to corresponding point clouds in the second point cloud image, to convert the charging port contour into a 3D object model; finding the charging port standard model in the 3D object model, and determining a model center of the charging port standard model in the 3D object model; determining a position corresponding to the model center as the target charging port position.

[0009] In some embodiments, after the target charging port position is obtained by performing charging port position identification processing on the target charging port according to the preset charging port standard model and the second point cloud image, the method further comprises: determining a target spatial position coordinate of the charging port relative to the charging robot arm according to a preset coordinate conversion relationship between the camera and the robot arm and the target charging port position; generating a motion path of the charging robot arm according to the target spatial position coordinate; controlling the charging robot arm to move to the target charging port position through the motion path to perform a charging operation on the target charging vehicle.

[0010] In some embodiments, before the first point cloud image of the target parking space corresponding area is obtained through the first camera, the method further comprises: obtaining a third point cloud image of the target parking space corresponding area through the first camera; determining a projection angle between a vehicle body of the target charging vehicle and a parking line of the target parking space through the third point cloud image; adjusting an image acquisition angle of the first camera according to the projection angle.

[0011] In some embodiments, the projection angle between the vehicle body of the target charging vehicle and the parking line of the target parking space is determined through the third point cloud image, comprising: performing cropping processing on the third point cloud image according to a depth dimension image in the third point cloud image to obtain a cropped third point cloud image; traversing point clouds in a lateral dimension image in the cropped third point cloud image to extract a plurality of second transition points; performing filtering processing on a plurality of the second transition points according to a preset second transition threshold interval to obtain a plurality of filtered second transition points; performing straight line fitting processing on the plurality of filtered second transition points to obtain a fitting straight line; mapping the fitting straight line into the third point cloud image to obtain a spatial fitting straight line, the spatial fitting straight line being the vehicle body of the target charging vehicle; recognizing the parking line in the third point cloud image; The projection angle between the space-fitting straight line and the parking space line is calculated by preset inverse perspective and projection principle.

[0012] In a second aspect, the embodiments of the present application further provide a charging port searching device, which is used for searching a target charging port of a target charging vehicle, the target charging vehicle is a to-be-charged vehicle parked in a target parking space, and a charging cover of the target charging vehicle is in an open state, and the charging port searching device comprises: a transceiving unit, configured to acquire a first point cloud image of a region corresponding to the target parking space through a first camera; a processing unit, configured to perform identification processing on the target charging vehicle in a charging port region through preset field potential method and the first point cloud image, to obtain a charging port region position corresponding to the charging port region, the charging port region comprising the charging cover and the target charging port; and move a charging mechanical arm to an image acquisition region corresponding to the charging port region position; The transceiving unit is further configured to acquire a second point cloud image through a second camera in the charging mechanical arm when the charging mechanical arm is in the image acquisition region. The processing unit is further configured to perform charging port position identification processing on the target charging port according to a preset charging port standard model and the second point cloud image, to obtain a target charging port position.

[0013] In a third aspect, the embodiments of the present application further provide a computer device, comprising a memory and a processor, the memory has stored a computer program, and the processor implements the above method when executing the computer program.

[0014] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, the storage medium stores a computer program, the computer program comprises program instructions, and the program instructions can implement the above method when executed by a processor.

[0015] The embodiment of the present application provides a charging port searching method, device, computer equipment and storage medium. The method is used for searching a target charging port of a target charging vehicle. The target charging vehicle is a to-be-charged vehicle parked in a target parking space, and a charging cover of the target charging vehicle is in an open state. The method comprises the following steps: acquiring a first point cloud image of a region corresponding to the target parking space through a first camera; performing charging port region identification processing on the target charging vehicle through a preset field potential method and the first point cloud image to obtain a charging port region position corresponding to the charging port region. The charging port region comprises the charging cover and the target charging port; moving a charging mechanical arm to an image acquisition region corresponding to the charging port region position; acquiring a second point cloud image through a second camera in the charging mechanical arm when the charging mechanical arm is in the image acquisition region; and performing charging port position identification processing on the target charging port according to a preset charging port standard model and the second point cloud image to obtain a target charging port position. The embodiment of the present application firstly performs rough positioning on the charging port through the field potential method to obtain the charging port region position, and then performs accurate positioning on the charging port based on the image re-acquired based on the charging port region position, so that the accurate charging port position is obtained. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0017] Figure 1 A flowchart of the charging port searching method provided by the embodiment of the present application is shown in the figure. Figure 2 An image diagram of the charging port searching process in the charging port searching method provided by the embodiment of the present application is shown in the figure. Figure 3 Another image diagram of the charging port searching process in the charging port searching method provided by the embodiment of the present application is shown in the figure. Figure 4 Another image diagram of the charging port searching process in the charging port searching method provided by the embodiment of the present application is shown in the figure. Figure 5 A sub-flowchart of the charging port searching method provided by the embodiment of the present application is shown in the figure. Figure 6 Another image diagram of the charging port searching process in the charging port searching method provided by the embodiment of the present application is shown in the figure. Figure 7 Another sub-flowchart of the charging port searching method provided by the embodiment of the present application is shown in the figure. Figure 8 Another image schematic diagram of the charging port searching process in the charging port searching method provided by the embodiment of the present application is shown in FIG. 4; Figure 9 Another image schematic diagram of the charging port searching process in the charging port searching method provided by the embodiment of the present application is shown in FIG. 4; Figure 10 A schematic block diagram of the charging port searching device provided by the embodiment of the present application is shown in FIG. 5; Figure 11 A schematic block diagram of the computer device provided by the embodiment of the present application is shown in FIG. 6. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort belong to the scope of protection of the present application.

[0019] It should be understood that, when used in the specification and the appended claims, the terms "comprise" and "include" indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0020] It should also be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and the appended claims of the present application, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] It should be further understood that the term "and / or" used in the specification and the appended claims of the present application means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0022] The embodiments of the present application provide a charging port searching method, device, computer device and storage medium.

[0023] The execution subject of the charging port searching method can be a charging port searching device provided by the embodiments of the present application, or a computer device integrated with the charging port searching device, wherein the charging port searching device can be realized in the form of hardware or software, and the computer device can be a terminal or a server.

[0024] Specifically, the computer device is in communication connection with the control of the first camera, the second camera and the charging mechanical arm. The first camera and the second camera can be the same camera, or can be different cameras. When they are different cameras, the second camera is arranged on the charging mechanical arm, and the first camera can be arranged on the charging mechanical arm or separated from the charging mechanical arm. At this time, it is arranged in the area where the target parking space image can be obtained. When the first camera is separated from the charging mechanical arm, the first camera can be responsible for the image acquisition of one or more charging parking spaces. At this time, the first camera is a large field of view 3D camera, which can observe objects within 0.3-20m. The second camera is a high-precision 3D camera.

[0025] When the first camera and the second camera are the same camera, the camera is arranged on the charging mechanical arm. At this time, the camera is a 3D camera that takes into account large field of view and high precision.

[0026] Among them, the charging port searching method provided by the application is used to search for a target charging port of a target charging vehicle. The target charging vehicle is a to-be-charged vehicle parked in a target parking space, and the charging cover of the target charging vehicle is in an open state. At this time, after the user parks the target charging vehicle in the target parking space and opens the charging cover, the computer device provided by the application can automatically identify the position of the target charging port in the target charging vehicle, so as to control the charging mechanical arm to aim the charging gun at the target charging vehicle based on the position of the target charging port, and perform automatic plug and unplug operations of the charging gun on the target charging vehicle. Without the user operating the charging gun, the problem of damaging the charging port due to improper manual operation and the safety hazard of electric shock due to manual operation are avoided.

[0027] Figure 1 is a flowchart of the charging port searching method provided by the embodiment of the application. As shown in Figure 1 , the method comprises the following steps S110-S150.

[0028] S110, acquiring a first point cloud image of a region corresponding to the target parking space by a first camera.

[0029] In this embodiment, after detecting that the target charging vehicle is parked in the target parking space and receiving a charging instruction for the target charging vehicle on the target parking space, the computer device acquires a first point cloud image of a region corresponding to the target parking space by a first camera.

[0030] In some embodiments, in order to improve the acquisition accuracy of the first point cloud image, the first camera of the present application first determines the angle between the vehicle and the parking line before acquiring the first point cloud image, and then adjusts the collection angle of the first camera based on the angle. When the first camera is located on the charging mechanical arm (such as the end of the mechanical arm), the angle of the charging mechanical arm is adjusted by adjusting the angle of the charging mechanical arm. Specifically, before step S110, the method further comprises: acquiring a third point cloud image of the target parking space corresponding region by the first camera; determining the projection angle between the body of the target charging vehicle and the parking line of the target parking space through the third point cloud image; and adjusting the image collection angle of the first camera according to the projection angle.

[0031] The projection angle between the body of the target charging vehicle and the parking line of the target parking space is determined through the third point cloud image, comprising: According to the depth dimension image in the third point cloud image, the third point cloud image is cropped to obtain a cropped third point cloud image; the point cloud in the transverse dimension image in the cropped third point cloud image is traversed to extract a plurality of second transition points; the plurality of second transition points are filtered according to a preset second transition threshold interval to obtain a plurality of filtered second transition points; the plurality of filtered second transition points are subjected to straight line fitting processing to obtain a fitting straight line; the fitting straight line is mapped into the third point cloud image to obtain a spatial fitting straight line, and the spatial fitting straight line is the body of the target charging vehicle; the parking line in the third point cloud image is identified; and the projection angle between the spatial fitting straight line and the parking line is calculated through a preset inverse perspective method and projection principle.

[0032] Specifically, a depth interval of the target charging vehicle in the target parking space is preset, and point cloud data outside the depth interval is deleted to obtain a cropped third point cloud image (the cropped image is shown in Figure 2 The transition data is traversed in the transverse dimension in the cropped third point cloud image corresponding region, the regional image gradient is calculated, and the transition points (as shown in Figure 3 , the red points are transition points) are extracted, then the high and low threshold values are set, the extracted points are filtered, the discrete points are filtered out, and the image edge is highlighted, then the least square method is used to fit a straight line (the fitting straight line is shown in Figure 4 ), the fitting straight line is mapped into the point cloud image to fit the straight line in space, and finally the inverse perspective method and the projection principle are combined to project the straight line in space and calculate the angle with the parking line (the projection angle).

[0033] The projection angle quantifies the degree of vehicle "parking deviation". Combined with the vehicle's projection angle and the known general vehicle size model, the three-dimensional spatial coordinate range of the "approximate charging port area" in the robot coordinate system can be further estimated, so as to adjust the position of the first camera and obtain a more accurate first point cloud image.

[0034] In other implementations, the degree of vehicle "parking deviation" (projection angle) can also be determined by judging the position of the front and rear wheels of the target charging vehicle. Specifically, the fourth point cloud image of the area corresponding to the target parking space is obtained by the first camera; the center points of the front and rear wheels on the same side of the target charging vehicle are determined by the fourth point cloud image, and the identified center points of the front and rear wheels are connected to obtain the actual parking axis of the vehicle. Then, the actual parking axis is compared with the standard direction of the charging pile connection line or the fixed reference object such as the suspension track to calculate the parking deviation angle (projection angle) of the vehicle.

[0035] The step of calculating the vehicle's "parking deviation" based on the position of the front and rear wheels can be performed after the "parking line" is not identified in the third point cloud image, or it can be performed directly (i.e., the calculation of the vehicle's "parking deviation" based on the "parking line" is not performed).

[0036] This solution relies entirely on the visual recognition capabilities of the first camera, without needing to depend on parking lines as a specific marker, thus ensuring accurate acquisition of the parking deflection angle of the target charging vehicle in various parking environments.

[0037] Furthermore, this application also acquires a fifth point cloud image through the first camera, and determines the distance between the target charging vehicle and adjacent vehicles or other obstacles based on the fifth point cloud image. If the distance is less than a preset safety threshold (e.g., 0.5 meters), subsequent operations are suspended and an alarm is issued to ensure the charging operation space of the robotic arm. Conversely, if the distance meets the requirements, the next operation is allowed, proceeding to step S120.

[0038] S120. The target charging vehicle is identified by using a preset field potential method and the first point cloud image to obtain the location of the charging port area corresponding to the charging port area. The charging port area includes the charging cover and the target charging port.

[0039] In some embodiments, such as Figure 5 As shown, step S120 includes: S1201. The first point cloud image is cropped according to the depth dimension image in the first point cloud image to obtain the cropped first point cloud image. S1202, Traverse the point clouds in the horizontal and vertical dimensions of the cropped first point cloud image to extract multiple first transition points. S1203. According to the preset first transition threshold range, filter out non-target points among multiple first transition points, and obtain multiple candidate transition point sets by clustering. S1204. Determine the discrete centers of the multiple candidate transition point sets; S1205. Calculate the distance between the transition center and the discrete center of each candidate transition point set; S1206. Based on the preset distance threshold and the distance values ​​corresponding to each of the candidate transition point sets, determine the target transition point set from the multiple candidate transition point sets, and determine the position of the target transition point set as the location of the charging port area.

[0040] The step of determining the target transition point set from multiple candidate transition point sets based on a preset distance threshold and the distance values ​​corresponding to each candidate transition point set includes: Filter candidate transition point sets from multiple candidate transition point sets whose distance values ​​are greater than the distance threshold to obtain a filtered candidate transition point set; if there are multiple filtered candidate transition point sets, the filtered candidate transition point set with the largest number of transition points in the multiple filtered candidate transition point sets is determined as the target transition point set; if there is only one filtered candidate transition point set, the filtered candidate transition point set is determined as the target transition point set.

[0041] Specifically, the acquired first point cloud image is decomposed into horizontal, vertical, and depth dimension images. Then, the first point cloud image is cropped based on the depth dimension image to reduce computational load. Specifically, the depth range of the target charging vehicle in the target parking space can be preset, and then point cloud data outside this depth range in the depth dimension image is deleted. Next, within the cropped area, the transition data is traversed along both the horizontal and vertical dimensions to extract transition points (e.g., ...). Figure 6 As shown, the colored dots around the charging cover are transition points. The final extracted transition points may also be the transition points corresponding to the charging port. This example only uses the transition points corresponding to the charging cover. Based on a pre-set first transition threshold range (e.g., the depth change of the transition point is less than 20cm), non-target points caused by factors such as tires are filtered out. Then, multiple candidate transition point sets are obtained by clustering, and the transition point sets are extracted. The transition center of each transition point set is extracted, and the distance between each transition center and the discrete center is calculated. Finally, high and low thresholds are set to filter discrete points and retain relatively concentrated point sets. The center of the point set region is calculated and determined as the center of the charging port region.

[0042] S130, move the charging robot arm to an image acquisition area corresponding to the charging port area position.

[0043] The image acquisition area is located near the charging port area position, and when the charging robot arm is in the image acquisition area, the second camera is directly opposite the charging port area position.

[0044] S140, when the charging robot arm is in the image acquisition area, acquiring a second point cloud image through the second camera in the charging robot arm.

[0045] When the charging robot arm is in the image acquisition area, the image of the charging port is acquired by the second camera, which improves the charging port recognition accuracy.

[0046] S150, according to the preset charging port standard model and the second point cloud image, performing charging port position recognition processing on the target charging port to obtain a target charging port position.

[0047] In some embodiments, as shown in Figure 7 S150 includes: S1501, determining a charging port center position in a depth dimension image of the second point cloud image according to a scalable 2D model of the charging port standard model and the depth dimension image in the second point cloud image; S1502, constructing a charging port target search area in the depth dimension image of the second point cloud image with the charging port center position as the center and with a preset area size; S1503, extracting a charging port contour from the charging port target search area according to a preset charging port depth threshold; S1504, mapping the charging port contour to a corresponding point cloud in the second point cloud image to convert the charging port contour into a 3D object model; S1505, searching for the charging port standard model in the 3D object model and determining a model center of the charging port standard model in the 3D object model; S1506, determining a position corresponding to the model center as the target charging port position.

[0048] Specifically, the charging port standard model is as shown in Figure 8 The model only retains the general features of various vehicle fast charging ports. Since full image search in 3D space is very time-consuming, this embodiment first performs fast and rough positioning on the 2D depth image (determining a charging port center position in a depth dimension image of the second point cloud image according to a scalable 2D model of the charging port standard model and the depth dimension image in the second point cloud image, as shown in Figure 9 Figure 9 ​The color part is the rough positioning matching charging port position, and the similarity (such as the correlation coefficient) of each position and each scaling ratio can be calculated, and then the position with the highest similarity and the scaling ratio are found, the center point of the position is the center coordinate of the charging port in the 2D image, that is, the center position of the charging port in the depth dimension image (2D center point), and then a region is drawn based on the 2D center point (for example, a cube region of interest with a side length of 5 cm is created). All subsequent 3D processing operations will only be performed on the point cloud in this region, rather than all point cloud data, thereby reducing the amount of calculation. Then, since the charging port is recessed, the depth value is different from the body, at this time, the depth threshold of the charging port can be set to extract the charging port contour, and then the points of the charging port contour extracted in the previous step are mapped from the 2D image coordinates back to their original 3D coordinates to form a 3D object model that only belongs to the charging port recessed part. Finally, the charging port standard model is searched (matched) in the generated 3D object model, and the position corresponding to the model center of the charging port standard model in the 3D object model is determined as the target charging port position.

[0049] After obtaining the target charging port position, a rigid affine transformation is required to convert the model center relative to the spatial coordinates of the robot arm, and the spatial coordinates are used to drive the charging robot arm to perform charging operation on the target charging vehicle. Specifically, after obtaining the target charging port position, the method further includes: determining the target space position coordinates of the charging port relative to the charging robot arm according to the preset coordinate conversion relationship between the camera and the robot arm and the target charging port position; generating a motion path of the charging robot arm according to the target space position coordinates; controlling the charging robot arm to move to the target charging port position through the motion path to perform charging operation on the target charging vehicle.

[0050] To sum up, the method provided by the embodiment of the present application is used for searching a target charging port of a target charging vehicle, the target charging vehicle is a to-be-charged vehicle parked in a target parking space, and a charging cover of the target charging vehicle is in an open state. The method comprises the following steps: acquiring a first point cloud image of a region corresponding to the target parking space through a first camera; performing identification processing on a charging port region of the target charging vehicle through a preset field potential method and the first point cloud image to obtain a charging port region position corresponding to the charging port region, the charging port region comprising the charging cover and the target charging port; moving a charging mechanical arm to an image acquisition region corresponding to the charging port region position; acquiring a second point cloud image through a second camera in the charging mechanical arm when the charging mechanical arm is in the image acquisition region; and performing charging port position identification processing on the target charging port according to a preset charging port standard model and the second point cloud image to obtain a target charging port position. The embodiment of the present application firstly performs coarse positioning on the charging port through the field potential method to obtain the charging port region position, and then performs accurate positioning on the charging port based on the image re-acquired based on the charging port region position, so as to obtain an accurate charging port position.

[0051] Figure 10 is a schematic block diagram of a charging port searching device provided by the embodiment of the present application. As shown in Figure 10 Corresponding to the above charging port searching method, the present application further provides a charging port searching device 1000. The charging port searching device 1000 comprises units for executing the above charging port searching method. The charging port searching device 1000 can be configured in a terminal or a server, and is used for searching a target charging port of a target charging vehicle, the target charging vehicle being a to-be-charged vehicle parked in a target parking space, and a charging cover of the target charging vehicle being in an open state. Specifically, please refer to Figure 10 The charging port searching device 1000 comprises a transceiver unit 1001 and a processing unit 1002, wherein: The transceiver unit 1001 is used for acquiring a first point cloud image of a region corresponding to the target parking space through a first camera. The processing unit 1002 is used for performing identification processing on a charging port region of the target charging vehicle through a preset field potential method and the first point cloud image to obtain a charging port region position corresponding to the charging port region, the charging port region comprising the charging cover and the target charging port; and moving a charging mechanical arm to an image acquisition region corresponding to the charging port region position. The transceiver unit 1001 is further used for acquiring a second point cloud image through a second camera in the charging mechanical arm when the charging mechanical arm is in the image acquisition region. The processing unit 1002 is further configured to perform charging port position identification processing on the target charging port according to a preset charging port standard model and the second point cloud image, to obtain a target charging port position.

[0052] In some embodiments, when performing the step of identifying the charging port area of the target charging vehicle according to the preset field potential method and the first point cloud image to obtain a charging port area position corresponding to the charging port area, the processing unit 1002 is specifically configured to: perform cropping processing on the first point cloud image according to a depth dimension image in the first point cloud image, to obtain a cropped first point cloud image; perform traversal on point clouds in a lateral dimension image and a longitudinal dimension image in the cropped first point cloud image, to extract a plurality of first transition points; filter non-target points in the plurality of first transition points according to a preset first transition threshold interval, and obtain a plurality of candidate transition point sets through clustering; determine discrete centers of the plurality of candidate transition point sets; respectively calculate distance values between transition centers of each candidate transition point set and the discrete centers; determine a target transition point set from the plurality of candidate transition point sets according to a preset distance threshold and the distance values respectively corresponding to each candidate transition point set, and determine a position of the target transition point set as the charging port area position.

[0053] In some embodiments, when performing the step of determining a target transition point set from the plurality of candidate transition point sets according to a preset distance threshold and the distance values respectively corresponding to each candidate transition point set, the processing unit 1002 is specifically configured to: filter candidate transition point sets with distance values greater than the distance threshold from the plurality of candidate transition point sets, to obtain filtered candidate transition point sets; if there are a plurality of filtered candidate transition point sets, determine a filtered candidate transition point set with the largest number of transition points from the plurality of filtered candidate transition point sets as the target transition point set; if there is only one filtered candidate transition point set, determine the filtered candidate transition point set as the target transition point set.

[0054] In some embodiments, when performing the step of performing charging port position identification processing on the target charging port according to a preset charging port standard model and the second point cloud image to obtain a target charging port position, the processing unit 1002 is specifically configured to: determine a charging port center position in a depth dimension image of the second point cloud image according to the scalable 2D model of the charging port standard model and the depth dimension image of the second point cloud image; construct a charging port target search area in the depth dimension image of the second point cloud image with the charging port center position as the center and with a preset area size; extract a charging port contour from the charging port target search area according to a preset charging port depth threshold; map the charging port contour to corresponding point clouds in the second point cloud image to convert the charging port contour into a 3D object model; find the charging port standard model in the 3D object model and determine a model center of the charging port standard model in the 3D object model; determine a position corresponding to the model center as the target charging port position.

[0055] In some embodiments, the processing unit 1002, after performing the step of performing charging port position identification processing on the target charging port according to the preset charging port standard model and the second point cloud image to obtain a target charging port position, is further configured to: determine a target spatial position coordinate of the charging port relative to the charging robot arm according to a preset coordinate conversion relationship between the camera and the robot arm and the target charging port position; generate a motion path of the charging robot arm according to the target spatial position coordinate; control the charging robot arm to move to the target charging port position through the motion path to perform a charging operation on the target charging vehicle.

[0056] In some embodiments, the transceiver unit 1001, before performing the step of acquiring a first point cloud image of a target parking space corresponding area through a first camera, is further configured to: acquire a third point cloud image of the target parking space corresponding area through the first camera; use the processing unit 1002 to determine a projection angle between a vehicle body of the target charging vehicle and a parking line of the target parking space through the third point cloud image; and adjust an image acquisition angle of the first camera according to the projection angle.

[0057] In some embodiments, the processing unit 1002, when performing the step of determining a projection angle between a vehicle body of the target charging vehicle and a parking line of the target parking space through the third point cloud image, is specifically configured to: perform cropping processing on the third point cloud image according to a depth dimension image in the third point cloud image to obtain a cropped third point cloud image; traverse the point cloud in the transverse dimension image in the cropped third point cloud image to extract a plurality of second transition points; filter the plurality of second transition points according to a preset second transition threshold interval to obtain a plurality of filtered second transition points; perform linear fitting processing on the plurality of filtered second transition points to obtain a fitted straight line; map the fitted straight line to the third point cloud image to obtain a spatial fitted straight line, the spatial fitted straight line being a vehicle body of the target charging vehicle; identify the parking space line in the third point cloud image; calculate the projection angle between the spatial fitted straight line and the parking space line through a preset inverse perspective method and projection principle.

[0058] To sum up, the embodiment of the application first coarsely searches the charging port through the field potential method to obtain the position of the charging port region, and then accurately positions the charging port based on the image reacquired based on the position of the charging port region, so as to obtain an accurate charging port position.

[0059] It should be noted that the specific implementation process of the charging port searching device 1000 and each unit can be clearly understood by those skilled in the art, and can refer to the corresponding description in the foregoing method embodiment. For the convenience and brevity of description, it will not be repeated here.

[0060] The charging port searching device 1000 described above can be implemented in the form of a computer program, which can run on a computer device as shown in the computer device. Figure 11 .

[0061] Please refer to Figure 11 , Figure 11 is a schematic block diagram of a computer device provided by the embodiment of the application. The computer device 1100 can be a terminal or a server, wherein the computer device 1100 is configured to search for a target charging port of a target charging vehicle, the target charging vehicle being a to-be-charged vehicle parked in a target parking space, and a charging cover of the target charging vehicle being in an open state.

[0062] Please refer to Figure 11 , the computer device 1100 includes a processor 1102, a memory and a network interface 1105 connected through a system bus 1101, wherein the memory can include a non-volatile storage medium 1103 and an internal memory 1104.

[0063] The non-volatile storage medium 1103 can store an operating system 11031 and a computer program 11032. The computer program 11032 includes program instructions that, when executed, cause the processor 1102 to perform a charging port search method.

[0064] The processor 1102 is configured to provide computing and control capabilities to support the operation of the entire computer device 1100.

[0065] The memory 1104 provides an environment for the execution of the computer program 11032 in the non-volatile storage medium 1103, which, when executed by the processor 1102, causes the processor 1102 to perform a charging port search method.

[0066] The network interface 1105 is configured to communicate with other devices over a network. Those skilled in the art can understand that the network interface 1105 can be configured to support wired communication or wireless communication, or both, and can be configured to support one or more communication technologies. Figure 11 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device 1100 to which the scheme of the present application is applied. The specific computer device 1100 can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0067] The processor 1102 is configured to run the computer program 11032 stored in the memory to implement the following steps: obtain a first point cloud image of the target parking space corresponding area through the first camera; perform a charging port area identification process on the target charging vehicle through a preset field potential method and the first point cloud image to obtain a charging port area position corresponding to the charging port area, the charging port area including the charging cover and the target charging port; move the charging robot arm to an image acquisition area corresponding to the charging port area position; obtain a second point cloud image through a second camera in the charging robot arm when the charging robot arm is in the image acquisition area; perform a charging port position identification process on the target charging port according to a preset charging port standard model and the second point cloud image to obtain a target charging port position.

[0068] It should be understood that, in the embodiments of the present application, the processor 1102 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0069] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiments can be completed by instructing relevant hardware by a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the above-mentioned embodiments.

[0070] Therefore, the present application also provides a storage medium. The storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program includes program instructions. The program instructions are executed by a processor to perform the following steps: obtaining a first point cloud image of the target parking space corresponding area through the first camera; performing identification processing on the charging port area of the target charging vehicle through a preset field potential method and the first point cloud image to obtain a charging port area position corresponding to the charging port area, wherein the charging port area includes the charging cover and the target charging port; moving the charging mechanical arm to an image acquisition area corresponding to the charging port area position; obtaining a second point cloud image through a second camera in the charging mechanical arm when the charging mechanical arm is in the image acquisition area; performing charging port position identification processing on the target charging port according to a preset charging port standard model and the second point cloud image to obtain a target charging port position.

[0071] The storage medium can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various computer-readable storage media that can store program codes.

[0072] Those skilled in the art can understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in general terms in the above description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0073] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of each unit is only a logical function division, and actual implementation can have another division manner. For example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed.

[0074] The steps in the method embodiments of the present application can be adjusted, combined and deleted in sequence according to actual needs. The units in the apparatus embodiments of the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0075] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a terminal or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application.

[0076] The above description is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for searching a charging port, characterized in that, The method is used to search for the target charging port of a target charging vehicle, wherein the target charging vehicle is a vehicle parked in a target parking space and awaiting charging, and the charging cover of the target charging vehicle is in an open state. The method includes: The first point cloud image of the area corresponding to the target parking space is obtained by the first camera; The target charging vehicle is identified by using a preset field potential method and the first point cloud image to obtain the location of the charging port area corresponding to the charging port area. The charging port area includes the charging cover and the target charging port. Move the charging robotic arm to the image acquisition area corresponding to the location of the charging port area; When the charging robotic arm is in the image acquisition area, a second point cloud image is acquired through the second camera in the charging robotic arm; The target charging port is identified by performing charging port location recognition processing based on the preset charging port standard model and the second point cloud image to obtain the target charging port location.

2. The method according to claim 1, characterized in that, The step of identifying the charging port region of the target charging vehicle using a preset field potential method and the first point cloud image to obtain the location of the charging port region includes: The first point cloud image is cropped based on the depth dimension image in the first point cloud image to obtain the cropped first point cloud image. The point clouds in the horizontal and vertical dimensions of the cropped first point cloud image are traversed to extract multiple first transition points. Based on a preset first transition threshold range, non-target points among multiple first transition points are filtered out, and multiple candidate transition point sets are obtained through clustering. Determine the discrete centers of the multiple candidate transition point sets; Calculate the distance between the transition center and the discrete center of each candidate transition point set; Based on a preset distance threshold and the distance values ​​corresponding to each candidate transition point set, a target transition point set is determined from multiple candidate transition point sets, and the location of the target transition point set is determined as the location of the charging port area.

3. The method according to claim 2, characterized in that, The step of determining the target transition point set from multiple candidate transition point sets based on a preset distance threshold and the distance values ​​corresponding to each candidate transition point set includes: Filter the candidate transition point sets from multiple candidate transition point sets whose distance values ​​are greater than the distance threshold to obtain a filtered candidate transition point set; If there are multiple filtered candidate transition point sets, then the filtered candidate transition point set with the largest number of transition points among the multiple filtered candidate transition point sets is determined as the target transition point set. If there is only one candidate transition point set after filtering, then the candidate transition point set after filtering is determined as the target transition point set.

4. The method according to claim 1, characterized in that, The step of performing charging port location identification processing on the target charging port based on a preset charging port standard model and the second point cloud image to obtain the target charging port location includes: The center position of the charging port in the depth dimension image of the second point cloud image is determined based on the scalable 2D model of the standard charging port model and the depth dimension image in the second point cloud image. Centered on the center position of the charging port, a target search area for the charging port is constructed in the depth dimension image of the second point cloud image with a preset area size; Based on a preset charging port depth threshold, the charging port outline is extracted from the charging port target search area. The charging port outline is mapped onto the corresponding point cloud in the second point cloud image to convert the charging port outline into a 3D object model. Locate the standard model of the charging port in the 3D object model, and determine the model center of the standard model of the charging port in the 3D object model; The location corresponding to the center of the model is determined as the target charging port location.

5. The method according to claim 1, characterized in that, After obtaining the target charging port location by performing charging port location identification processing based on a preset charging port standard model and the second point cloud image, the method further includes: Based on the preset coordinate transformation relationship between the camera and the robotic arm and the target charging port position, the target spatial position coordinates of the charging port relative to the charging robotic arm are determined. The motion path of the charging robotic arm is generated based on the target spatial coordinates. The charging robotic arm is controlled to move along the motion path to the target charging port position to perform a charging operation on the target charging vehicle.

6. The method according to any one of claims 1 to 5, characterized in that, Before acquiring the first point cloud image of the area corresponding to the target parking space via the first camera, the method further includes: The third point cloud image of the area corresponding to the target parking space is obtained through the first camera; The projection angle between the body of the target charging vehicle and the parking line of the target parking space is determined by the third point cloud image. Adjust the image acquisition angle of the first camera according to the projection angle.

7. The method according to claim 6, characterized in that, Determining the projection angle between the body of the target charging vehicle and the parking line of the target parking space using the third point cloud image includes: The third point cloud image is cropped based on the depth dimension image in the third point cloud image to obtain the cropped third point cloud image. The point cloud in the horizontal dimension of the cropped third point cloud image is traversed to extract multiple second transition points; Based on the preset second transition threshold range, multiple second transition points are filtered to obtain multiple filtered second transition points. A straight line is fitted to multiple filtered second transition points to obtain a fitted straight line. The fitted line is mapped onto the third point cloud image to obtain a spatial fitted line, which is the body of the target charging vehicle; Identify the parking space lines in the third point cloud image; The projection angle between the spatial fitting line and the parking space line is calculated using a preset inverse perspective method and projection principle.

8. A charging port search device, characterized in that, The charging port search device is used to search for the target charging port of a target charging vehicle, wherein the target charging vehicle is a vehicle parked in a target parking space and waiting to be charged, and the charging port cover of the target charging vehicle is in an open state. The charging port search device includes: The transceiver unit is used to acquire a first point cloud image of the area corresponding to the target parking space through the first camera; The processing unit is used to identify the charging port area of ​​the target charging vehicle using a preset field potential method and the first point cloud image, to obtain the charging port area position corresponding to the charging port area, wherein the charging port area includes the charging cover and the target charging port; and to move the charging robotic arm to the image acquisition area corresponding to the charging port area position. The transceiver unit is also used to acquire a second point cloud image through a second camera in the charging robotic arm when the charging robotic arm is in the image acquisition area; The processing unit is further configured to perform charging port location recognition processing on the target charging port according to the preset charging port standard model and the second point cloud image, so as to obtain the target charging port location.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the charging port search method as described in any one of claims 1-7.

10. A storage medium, characterized in that, The storage medium stores a computer program, which includes program instructions that, when executed by a processor, cause the processor to perform the charging port search method as described in any one of claims 1-7.

Images