Parking space adjusting method, electronic equipment, storage medium and program product

Abstract

The invention provides a parking space adjustment method, electronic equipment, a storage medium and a program product, and relates to the technical field of automatic driving. The method comprises the following steps: acquiring a space parking space frame and obstacle data generated when a vehicle is in a parking environment without a parking space line; according to the spatial parking space frame and the obstacle data, based on a preset strategy, determining an invasion state of the obstacle to the spatial parking space frame; and according to the invasion state and the obstacle data, adjusting the space parking space frame to obtain a target space parking space frame for guiding the vehicle to park. Thus, the parking space frame is generated based on the environment without the parking space line, the invasion state of the obstacle to the parking space frame is judged based on the obstacle data, then the parking space frame is adjusted based on the invasion state and the obstacle data, and the situation that the parking space frame is limited by the receptive field of the vehicle sensor is avoided. Therefore, the risk that the vehicle collides when the vehicle is parked in the parking space frame exists. The problems that a traditional parking space division mode is insufficient in flexibility, and a generated parking space frame cannot be flexibly adjusted are solved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a parking space adjustment method, electronic equipment, storage medium and program product. Background Art

[0002] With the popularization of new energy vehicles, intelligence has an increasingly important position in the configuration of automobiles. Among the current intelligent applications of automobiles, intelligent parking is undoubtedly one of the important functions. Intelligent parking requires the vehicle to identify the parking location in the environment through sensors, which is generally a marked parking space or a parking area. The recognition scheme for marked parking spaces is now very mature. Through the on-board camera and deep learning methods, the parking space lines drawn in the environment can be recognized more accurately. For the division of unmarked space parking spaces, it is usually based on the area occupied by obstacles. The connected area between two adjacent obstacles with a spatial position greater than the length and width of the vehicle body is divided as a parking space. The parking space generated by this method is limited by the surround view image. In actual applications, the surround view image obtained by the vehicle during parking is dynamic, and obstacles that were not detected during the parking space division process may appear in the parking space.

[0003] Therefore, although the existing parking space detection / division method can generate spatial parking spaces that are not based on parking lines, it is not flexible enough and cannot flexibly adjust the generated parking space frame. Summary of the invention

[0004] In view of this, the purpose of the embodiments of the present application is to provide a parking space adjustment method, electronic device, storage medium and program product, which can improve the problem that although the traditional parking space division method can generate spatial parking spaces that are not based on parking lines, it lacks flexibility and cannot flexibly adjust the generated parking space frame.

[0005] In order to achieve the above technical objectives, the technical solutions adopted in this application are as follows:

[0006] In a first aspect, an embodiment of the present application provides a parking space adjustment method, the method comprising:

[0007] Obtain the spatial parking space frame and obstacle data generated when the vehicle is in a parking environment without parking space lines;

[0008] According to the space parking space frame and the obstacle data, based on a preset strategy, determining the intrusion state of the obstacle into the space parking space frame;

[0009] The spatial parking space frame is adjusted according to the intrusion state and the obstacle data to obtain a target spatial parking space frame for guiding the parking of the vehicle.

[0010] In combination with the first aspect, in some optional implementations, the obstacle data includes a circumscribed rectangular placeholder frame and point cloud data of the obstacle;

[0011] The determining, based on the spatial parking space frame and the obstacle data and a preset strategy, of an intrusion state of an obstacle into the spatial parking space frame includes:

[0012] Determining an intersection-and-joint ratio of the spatial parking space frame to the circumscribed rectangular placeholder frame according to the spatial parking space frame and the circumscribed rectangular placeholder frame;

[0013] When the intersection-over-union ratio is less than a first preset threshold, the intrusion state is determined according to the spatial parking space frame and the point cloud data.

[0014] In combination with the first aspect, in some optional implementations, the point cloud data includes a laser point cloud mapped to the same plane as the parking space frame and first two-dimensional coordinates corresponding to pixel points in the laser point cloud;

[0015] When the intersection-over-union ratio is less than a first preset threshold, determining the intrusion state according to the spatial parking space frame and the point cloud data includes:

[0016] When the intersection-over-union ratio is less than the first preset threshold, determining the number of first target pixel points according to the spatial parking space frame and the first two-dimensional coordinates, where the first target pixel points represent the pixel points whose first two-dimensional coordinates are located in the spatial parking space frame;

[0017] When the number of the first target pixel points is greater than or equal to a second preset threshold, it is determined that the intrusion state is a first intrusion state indicating that an obstacle is contained in the space parking space frame.

[0018] In combination with the first aspect, in some optional implementations, the point cloud data includes a second two-dimensional coordinate corresponding to a pixel point in a surround view point cloud and a pixel point in the surround view point cloud mapped to the same plane as the parking space frame;

[0019] When the intersection-over-union ratio is less than a first preset threshold, determining the intrusion state according to the spatial parking space frame and the point cloud data includes:

[0020] When the intersection-over-union ratio is less than the first preset threshold, determining the number of second target pixel points according to the spatial parking space frame and the second two-dimensional coordinates, where the second target pixel points represent the pixel points whose second two-dimensional coordinates are located in the spatial parking space frame;

[0021] When the number of the second target pixel points is greater than or equal to a third preset threshold, determining a covariance matrix corresponding to the second target pixel points according to the coordinates of the second target pixel points;

[0022] Performing eigenvalue decomposition on the covariance matrix to obtain the largest eigenvalue and the second largest eigenvalue;

[0023] When the magnitudes of the maximum eigenvalue and the second largest eigenvalue meet a preset condition, it is determined that the intrusion state is a first intrusion state indicating that an obstacle is contained in the space parking space frame.

[0024] In combination with the first aspect, in some optional implementations, when the intrusion state is the first intrusion state, adjusting the spatial parking space frame according to the intrusion state and the obstacle data to obtain a target spatial parking space frame for guiding parking of the vehicle includes:

[0025] When the intrusion state is the first intrusion state, determining the intrusion direction and intrusion distance corresponding to each pixel point according to the coordinates corresponding to each pixel point in the point cloud data;

[0026] Taking the set of the pixel points with the same invasion direction as a first unit set, and for each first unit set, determining a first average invasion distance of all the pixel points in the first unit set;

[0027] According to the intrusion direction corresponding to all the pixel points in the first unit set and the first average intrusion distance, the spatial parking space frame is adjusted to obtain the target spatial parking space frame, wherein the adjustment direction of the spatial parking space frame is the opposite direction of the intrusion direction corresponding to all the pixel points in the first unit set, and the adjustment distance of the spatial parking space frame is the first average intrusion distance plus a preset tolerance gap.

[0028] In combination with the first aspect, in some optional implementations, the obstacle data further includes a contour boundary line of the obstacle;

[0029] When the intrusion state is the first intrusion state, adjusting the space parking space frame according to the intrusion state and the obstacle data to obtain a target space parking space frame for guiding the parking of the vehicle includes:

[0030] When the intrusion state is the first intrusion state, determining, in the contour boundary line, a line segment having a distance from the space parking space frame less than a fourth preset threshold, an angle less than a preset angle and the longest length as a reference boundary line;

[0031] Adjusting the direction of the space parking space frame so that the space parking space frame is parallel to the reference boundary line;

[0032] Adjusting the space parking space frame according to the reference boundary line to obtain an adjusted space parking space frame;

[0033] The adjusted space parking space frame is adjusted according to other line segments in the contour boundary line except the reference boundary line to obtain the target space parking space frame.

[0034] In combination with the first aspect, in some optional implementations, adjusting the space parking space frame according to the reference boundary line to obtain the adjusted space parking space frame includes:

[0035] When there are other vehicles in front of the spatial parking space frame, and the minimum distance between the circumscribed rectangular placeholder frame corresponding to the other vehicles and the reference boundary line is less than a fifth preset threshold, the direction of the spatial parking space frame is adjusted so that the distance between the side of the spatial parking space frame closest to the reference boundary line and the reference boundary line is less than a sixth preset threshold, to obtain an adjusted spatial parking space frame.

[0036] In combination with the first aspect, in some optional implementations, the adjusting the space parking space frame according to the reference boundary line to obtain the adjusted space parking space frame further includes:

[0037] When there are other vehicles in front of the spatial parking space frame, and the minimum distance between the circumscribed rectangular placeholder frame corresponding to the other vehicles and the reference boundary line is greater than or equal to a fifth preset threshold, the position of the spatial parking space frame is adjusted so that a line connecting the center point of the circumscribed rectangular placeholder frame corresponding to the other vehicles and the center point of the spatial parking space frame is parallel to the reference boundary line, thereby obtaining the adjusted spatial parking space frame.

[0038] In combination with the first aspect, in some optional implementations, adjusting the adjusted space parking space frame to obtain the target space parking space frame according to other line segments in the contour boundary line except the reference boundary line includes:

[0039] The other line segments in the contour boundary line except the reference boundary line are taken as a modified line segment set, and the intrusion direction and intrusion distance corresponding to the third target pixel point in the modified line segment set are determined according to the coordinates corresponding to the third target pixel point in the modified line segment set, wherein the third target pixel point represents the endpoint and the midpoint of each line segment in the modified line segment set located in the adjusted spatial parking space frame;

[0040] Taking the set of the third target pixel points with the same invasion direction as a second unit set, and for each second unit set, determining a second average invasion distance of all the third target pixel points in the second unit set;

[0041] According to the invasion direction corresponding to all the third target pixel points in the second unit set and the second average invasion distance, the adjusted spatial parking space frame is adjusted to obtain the target spatial parking space frame, wherein the adjustment direction of the adjusted spatial parking space frame is the opposite direction of the invasion direction corresponding to all the pixel points in the second unit set, and the adjustment distance of the adjusted spatial parking space frame is the second average invasion distance plus the second preset tolerance gap.

[0042] In combination with the first aspect, in some optional implementations, between acquiring the spatial parking space frame and obstacle data generated when the vehicle is in a parking environment without parking space lines, and determining the intrusion state of the obstacle into the spatial parking space frame based on a preset strategy according to the spatial parking space frame and the obstacle data, the method further includes:

[0043] Preprocessing the obstacle data to perform time synchronization and format conversion on the obstacle data to obtain preprocessed obstacle data;

[0044] The determining, based on the spatial parking space frame and the obstacle data and a preset strategy, of an intrusion state of an obstacle into the spatial parking space frame includes:

[0045] According to the spatial parking space frame and the pre-processed obstacle data, based on a preset strategy, determining an intrusion state of the obstacle into the spatial parking space frame;

[0046] The step of adjusting the space parking space frame according to the intrusion state and the obstacle data to obtain a target space parking space frame for guiding the vehicle to park includes:

[0047] The spatial parking space frame is adjusted according to the intrusion state and the pre-processed obstacle data to obtain a target spatial parking space frame for guiding the parking of the vehicle.

[0048] In a second aspect, an embodiment of the present application further provides an electronic device, comprising a processor and a memory coupled to each other, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the electronic device executes the above method.

[0049] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium. When the computer program is executed on a computer, the computer executes the above method.

[0050] In a fourth aspect, an embodiment of the present application further provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0051] The invention adopting the above technical solution has the following advantages:

[0052] In the technical solution provided in the present application, the spatial parking space frame and obstacle data generated when the vehicle is in a parking environment without parking space lines are first obtained. Then, according to the spatial parking space frame and the obstacle data, based on the preset strategy, the intrusion state of the obstacle to the spatial parking space frame is determined. Finally, according to the intrusion state and the obstacle data, the spatial parking space frame is adjusted to obtain the target spatial parking space frame for guiding the parking of the vehicle. In this way, a spatial parking space is generated based on an environment without parking space lines, and the intrusion state of the obstacles around the vehicle to the spatial parking space is judged based on the obstacle data, and then the spatial parking space is adjusted based on the intrusion state and the obstacle data, so as to avoid the generated spatial parking space being limited by the receptive field of the vehicle sensor, so that new obstacles appear in the spatial parking space during the parking process of the vehicle, and there is a risk of collision between the vehicles. Although the traditional parking space division method can generate spatial parking spaces that are not based on parking lines, it is not flexible enough and cannot flexibly adjust the generated parking space frame. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The present application may be further described by the non-limiting embodiments given in the accompanying drawings. It should be understood that the following drawings only illustrate certain embodiments of the present application and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings may be obtained based on these drawings without creative effort.

[0054] Figure 1 A structural block diagram of an electronic device provided in an embodiment of the present application.

[0055] Figure 2 A flowchart of a parking space adjustment method provided in an embodiment of the present application.

[0056] Figure 3 A schematic diagram of the occupancy of the circumscribed rectangular placeholder frame provided in an embodiment of the present application.

[0057] Figure 4 A schematic diagram of point cloud data occupancy provided in an embodiment of the present application.

[0058] Figure 5 A schematic diagram of contour boundary line occupancy provided for an embodiment of the present application.

[0059] Figure 6 A schematic diagram of a parking space frame after adjustment based on a reference boundary line provided in an embodiment of the present application.

[0060] Figure 7 A schematic diagram of the complete execution flow of the parking space adjustment method provided in an embodiment of the present application.

[0061] Icon: 100 - electronic device; 101 - processor; 102 - memory. DETAILED DESCRIPTION

[0062] The present application will be described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that in the drawings or descriptions, similar or identical parts use the same figure numbers, and the implementation methods not shown or described in the drawings are forms known to ordinary technicians in the relevant technical field. In the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0063] Please refer to Figure 1 , an embodiment of the present application provides an electronic device 100 that may include a processor 101 and a memory 102. The memory 102 stores a computer program, and when the computer program is executed by the processor 101, the electronic device 100 can perform the corresponding steps in the following parking space adjustment method.

[0064] In this embodiment, the electronic device 100 may be a personal computer, a laptop computer, a vehicle-mounted central controller, etc. It is used to obtain the space parking space frame and obstacle data generated when the vehicle is in a parking environment without parking space lines. Then, according to the space parking space frame and the obstacle data, based on a preset strategy, the intrusion state of the obstacle to the space parking space frame is determined. Finally, according to the intrusion state and the obstacle data, the space parking space frame is adjusted to obtain a target space parking space frame for guiding the vehicle to park.

[0065] In this embodiment, the processor 101 may be an integrated circuit chip having signal processing capabilities. The processor 101 may be a general-purpose processor. For example, the processor 101 may be a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and may implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present application.

[0066] The memory 102 may be, but is not limited to, a random access memory, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, etc. In this embodiment, the memory 102 may be used to store a space parking space frame, obstacle data, a preset strategy, an intrusion state, a target space parking space frame, etc. Of course, the memory 102 may also be used to store a program, and the processor 101 executes the program after receiving an execution instruction.

[0067] Understandably, Figure 1 The structure of the electronic device 100 shown in FIG. 1 is only a schematic diagram of a structure. The electronic device 100 may also include Figure 1 More components are shown. Figure 1 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0068] Please refer to Figure 2 The present application also provides a parking space adjustment method, which can be applied to the above electronic device 100, and each step in the method is executed or implemented by the electronic device 100. The parking space adjustment method can include the following steps:

[0069] Step 210, obtaining the spatial parking space frame and obstacle data generated when the vehicle is in a parking environment without parking space lines;

[0070] Step 220, determining the intrusion state of the obstacle into the space parking space frame based on the preset strategy according to the space parking space frame and the obstacle data;

[0071] Step 230: adjusting the spatial parking space frame according to the intrusion status and the obstacle data to obtain a target spatial parking space frame for guiding the parking of the vehicle.

[0072] In the above-mentioned implementation, the spatial parking space frame and obstacle data generated when the vehicle is in a parking environment without parking space lines are first obtained. Then, according to the spatial parking space frame and the obstacle data, based on a preset strategy, the intrusion state of the obstacle into the spatial parking space frame is determined. Finally, according to the intrusion state and the obstacle data, the spatial parking space frame is adjusted to obtain a target spatial parking space frame for guiding the parking of the vehicle. In this way, a spatial parking space is generated based on an environment without parking space lines, and the intrusion state of the obstacles around the vehicle into the spatial parking space is judged based on the obstacle data, and then the spatial parking space is adjusted based on the intrusion state and the obstacle data, so as to avoid the generated spatial parking space being limited by the receptive field of the vehicle sensor, so that new obstacles appear in the spatial parking space during the parking process of the vehicle, and there is a risk of collision between the vehicles. Although the traditional parking space division method can generate spatial parking spaces that are not based on parking lines, it is not flexible enough and cannot flexibly adjust the generated parking space frame.

[0073] The following will explain in detail the steps of the parking space adjustment method, as follows:

[0074] In step 210, the spatial parking space frame may be a parking space pre-generated when the vehicle is in a parking environment without parking space lines. The parking space may be a connected area between the circumscribed rectangles of two adjacent obstacles obtained by obtaining an environmental image of the vehicle's surroundings through a sensor mounted on the vehicle and identifying the environmental image. The spatial parking space frame may be represented by a proportionally scaled rectangle of the size of a standard parking space.

[0075] In this embodiment, the obstacle data may include laser fs (freespace) of obstacles near the vehicle: that is, the point cloud data of the obstacle detected by the laser radar carried by the vehicle, and the point cloud data is mapped to the same plane of the spatial parking space frame (from the three-dimensional point cloud with coordinates (x, y, z) to the two-dimensional point with coordinates (x, y)) to obtain a laser point cloud; surround view fs: that is, the point cloud data of the obstacle detected by the surround view fisheye carried by the vehicle, and the point cloud data is mapped to the same plane of the spatial parking space frame to obtain a surround view point cloud; laser OD (Object Detection): that is, the obstacle detection / recognition result (expressed as a three-dimensional frame or a two-dimensional frame) detected by the laser radar carried by the vehicle, and the three-dimensional frame in the obstacle detection / recognition result is mapped to the same plane of the spatial parking space frame to obtain the circumscribed rectangular placeholder frame of the obstacle; ultrasonic cluster: that is, the clustering result of the obstacle contour / boundary point detected by the ultrasonic radar, which is expressed as multiple line segments, that is, the contour boundary line.

[0076] It is understandable that the obstacle data may also include the coordinates of the pixel points constituting the above-mentioned laser point cloud, surround point cloud, circumscribed rectangular placeholder frame and contour boundary line (the coordinates may be based on the center of the vehicle as the origin). The obstacle proposed in this embodiment is a relatively broad general term, that is, all obstacles that may invade the space parking space frame and cause a collision risk for the parking vehicle are regarded as obstacles, including but not limited to drivable lanes, curbs, fixed obstacles, mobile obstacles, pedestrians, etc.

[0077] In this embodiment, the acquisition of the spatial parking space frame and obstacle data may be that the spatial parking space frame and obstacle data are generated in real time according to the surrounding environment of the vehicle during the driving process of the vehicle, and the processor 101 in the electronic device 100 calls them in real time and performs subsequent intrusion state judgment and spatial parking space frame adjustment based on the spatial parking space frame and obstacle data; or, the acquisition of the spatial parking space frame and obstacle data may also be that in the vehicle development stage, the user customizes the surrounding environment of the vehicle, and customizes the spatial parking space frame and obstacle data generated in the surrounding environment based on the surrounding environment, and stores them in the memory 102 in the electronic device 100, so that they can be called by the processor 101 of the electronic device 100 during the development and testing process of subsequent intrusion state judgment and spatial parking space frame adjustment. The acquisition method of the spatial parking space frame and obstacle data is not specifically limited here.

[0078] Between step 210 and step 220, the method may further include:

[0079] The obstacle data is preprocessed to perform time synchronization and format conversion on the obstacle data to obtain preprocessed obstacle data.

[0080] In this embodiment, after obtaining the obstacle data, the laser point cloud, surround point cloud, circumscribed rectangular placeholder box and contour boundary line in the obstacle data are time synchronized, and the contour boundary line is converted into a local data format (including the two endpoints and center point coordinates of each line segment in the contour boundary line) to obtain the preprocessed obstacle data.

[0081] It is understandable that the difference between the preprocessed laser point cloud, surround point cloud, circumscribed rectangular placeholder frame and contour boundary line and the obstacle data in step 210 is only the time arrangement order and format. The role of preprocessing is to improve the convenience of obstacle data related calculations in the subsequent intrusion state judgment and space parking frame adjustment process, and does not affect the calculation and processing logic involved in the subsequent processing process. Therefore, for ease of understanding, the obstacle data is still used as the data basis in the subsequent intrusion state and space parking frame adjustment process. In the actual application of intrusion state judgment and space parking frame adjustment, the preprocessed obstacle data can be replaced by the preprocessed data.

[0082] In step 220, determining the intrusion state of the obstacle into the spatial parking space frame based on the spatial parking space frame and the obstacle data and based on a preset strategy may include:

[0083] Determining an intersection-and-joint ratio of the spatial parking space frame to the circumscribed rectangular placeholder frame according to the spatial parking space frame and the circumscribed rectangular placeholder frame;

[0084] When the intersection-over-union ratio is less than a first preset threshold, the intrusion state is determined according to the spatial parking space frame and the point cloud data.

[0085] In this embodiment, refer to Figure 3 , slot represents the space parking space frame, od represents the circumscribed rectangular placeholder frame of the obstacle, and IOU represents the intersection-over-union ratio of the two. In this embodiment, the intersection-over-union ratio of the space parking space frame and the circumscribed rectangular placeholder frame = the area of ​​the intersection of the two / (the total area of ​​the two-the overlapping area).

[0086] In this embodiment, the first preset threshold can be flexibly set according to user needs, such as 1 / 8, 1 / 10, 1 / 12, etc.

[0087] It is understandable that when the intersection and union ratio of the spatial parking space frame and the circumscribed rectangular placeholder frame of the obstacle is greater than or equal to the first preset threshold, it is determined that the pre-generated spatial parking space frame is occupied, and the parking space is set to a non-adjustable state (i.e., a non-parking state).

[0088] In this embodiment, the point cloud data includes a laser point cloud mapped to the same plane as the parking space frame and first two-dimensional coordinates corresponding to pixel points in the laser point cloud;

[0089] When the intersection-over-union ratio is less than a first preset threshold, determining the intrusion state according to the spatial parking space frame and the point cloud data may include:

[0090] When the intersection-over-union ratio is less than the first preset threshold, determining the number of first target pixel points according to the spatial parking space frame and the first two-dimensional coordinates, where the first target pixel points represent the pixel points whose first two-dimensional coordinates are located in the spatial parking space frame;

[0091] When the number of the first target pixel points is greater than or equal to a second preset threshold, it is determined that the intrusion state is a first intrusion state indicating that an obstacle is contained in the space parking space frame.

[0092] In this embodiment, refer to Figure 4 , fs represents point cloud data, i.e., laser point cloud or surround point cloud, i.e., the black dots in the figure. Among them, when the coordinates of all the pixel points constituting the spatial parking space frame and the coordinates of the point cloud data are known, it is a conventional technical means in the field of geometry to determine the relative position of the pixel points in each point cloud data relative to the spatial parking space frame by each coordinate, which will not be elaborated here.

[0093] In this embodiment, when the number of pixel points (i.e., the first target pixel points) of the laser point cloud located within the spatial parking space frame is greater than or equal to a second preset threshold (flexibly set according to user needs, such as 20, 30, 50, etc.), it is determined that the spatial parking space frame is occupied by fs, and the spatial parking space frame is set to an occupied state (i.e., the first intrusion state).

[0094] In this embodiment, the point cloud data includes a second two-dimensional coordinate corresponding to a pixel point in a surround view point cloud and a surround view point cloud mapped to the same plane as the parking space frame;

[0095] When the intersection-over-union ratio is less than a first preset threshold, determining the intrusion state according to the spatial parking space frame and the point cloud data may further include:

[0096] When the intersection-over-union ratio is less than the first preset threshold, determining the number of second target pixel points according to the spatial parking space frame and the second two-dimensional coordinates, where the second target pixel points represent the pixel points whose second two-dimensional coordinates are located in the spatial parking space frame;

[0097] When the number of the second target pixel points is greater than or equal to a third preset threshold, determining a covariance matrix corresponding to the second target pixel points according to the coordinates of the second target pixel points;

[0098] Performing eigenvalue decomposition on the covariance matrix to obtain the largest eigenvalue and the second largest eigenvalue;

[0099] When the magnitudes of the maximum eigenvalue and the second largest eigenvalue meet a preset condition, it is determined that the intrusion state is a first intrusion state indicating that an obstacle is contained in the space parking space frame.

[0100] In this embodiment, the third preset threshold can be flexibly set according to user needs, such as 20, 30, 50, etc.

[0101] In this embodiment, the surround view point cloud can be segmented for each pixel in the surround view point cloud according to the recognition result of the vehicle to obtain categories including but not limited to drivable lanes, curbs, fixed obstacles, mobile obstacles, etc. The surround view point cloud in this embodiment takes the pixel points of the curb category as an example.

[0102] In this embodiment, the number of pixels (i.e., second target pixels) of the view point cloud located in the space parking space frame is determined according to the space parking space frame and the second two-dimensional coordinates. When the number is greater than or equal to the third preset threshold, the covariance matrix of the coordinates of the second target pixel points is determined according to the coordinates of the second target pixel points (conventional technical means, which will not be described here). Then, the covariance matrix is ​​eigen-decomposed to obtain the maximum eigenvalue and the second largest eigenvalue. When the sizes of the two meet the preset conditions, the space parking space frame is set to an occupied state.

[0103] In this embodiment, the preset condition can be flexibly set according to user needs. The preset condition in this embodiment can be that when the maximum eigenvalue is much larger than the second largest eigenvalue (the judgment standard of "much larger than" can be flexibly set according to user needs, for example, the maximum eigenvalue can be 5 times and above, 8 times and above, 10 times and above, etc. of the second largest eigenvalue), the intrusion state of the space parking space frame is set to the occupied state, that is, the first intrusion state.

[0104] It is understandable that when the number of pixels of the laser point cloud located in the spatial parking space frame is less than the second preset threshold, and the maximum eigenvalue of the covariance matrix corresponding to the surround point cloud is far less than the second largest eigenvalue, the intrusion state of the spatial parking space frame is set to the second intrusion state that characterizes that there are no obstacles in the parking space frame, that is, the unoccupied state. At this time, there is no need to make subsequent adjustments to the spatial parking space frame, and the vehicle can be directly controlled to park in the spatial parking space frame.

[0105] In step 230, when the intrusion state is the first intrusion state, adjusting the space parking space frame according to the intrusion state and the obstacle data to obtain a target space parking space frame for guiding the parking of the vehicle may include:

[0106] When the intrusion state is the first intrusion state, determining the intrusion direction and intrusion distance corresponding to each pixel point according to the coordinates corresponding to each pixel point in the point cloud data;

[0107] Taking the set of the pixel points with the same invasion direction as a first unit set, and for each first unit set, determining a first average invasion distance of all the pixel points in the first unit set;

[0108] According to the intrusion direction corresponding to all the pixel points in the first unit set and the first average intrusion distance, the spatial parking space frame is adjusted to obtain the target spatial parking space frame, wherein the adjustment direction of the spatial parking space frame is the opposite direction of the intrusion direction corresponding to all the pixel points in the first unit set, and the adjustment distance of the spatial parking space frame is the first average intrusion distance plus a first preset tolerance gap.

[0109] In this embodiment, when the spatial parking space frame is occupied by an obstacle (i.e., the intrusion state is the first intrusion state), the intrusion direction and intrusion distance of each pixel point in the spatial parking space frame are calculated (for pixels not in the spatial parking space frame, the intrusion direction and intrusion distance are directly set to null values). The intrusion distance may refer to the distance from the current pixel point to each edge of the spatial parking space frame, and the intrusion direction may be the direction of the edge of the spatial parking space frame that is closest to the current pixel point (for example, if the pixel point is 30 pixels from the left side of the spatial parking space frame, 500 pixels from the right side, 900 pixels from the upper side, and 1200 pixels from the lower side, the intrusion direction is the closest left side).

[0110] In this embodiment, please refer to Figure 4 a. For pixel points whose invasion distance is less than the seventh preset threshold (flexibly set according to user needs, such as 30 pixels, 50 pixels, 200 pixels, etc., and 200 pixels is taken as an example in this embodiment) and are not empty, they are determined as adjustment reference points. For pixel points whose invasion distance is greater than the seventh preset threshold and are not empty, they are determined as non-adjustable points ( Figure 4 Non-adjustable points are not shown).

[0111] In this embodiment, the pixels with the same intrusion direction and not empty are classified into a first unit set, including a left intrusion set, a right intrusion set, an upper intrusion set, and a lower intrusion set. For each intrusion set, the first average intrusion distance of all the pixels therein is calculated (that is, the average of the invasion distances of all the pixels), and then the spatial parking space frame is adjusted according to the invasion direction and the first average intrusion distance to obtain the target spatial parking space frame. Specifically, the adjustment direction of the spatial parking space frame is the opposite direction of the invasion direction of each pixel (for example, when the invasion direction is to the left, the adjustment direction is to the right), and the adjustment distance of the spatial parking space frame is the first average intrusion distance plus a preset tolerance gap. Please refer to Figure 4 b, is the effect after the spatial parking space frame is adjusted according to the adjustment direction and adjustment distance.

[0112] In this embodiment, the preset tolerance gap can be flexibly set according to user needs, such as 10 pixels, 50 pixels, 100 pixels, etc.

[0113] In this embodiment, please refer to Figure 4 c. When the number of unadjustable points is greater than or equal to the eighth preset threshold (flexibly set according to user needs, such as 30, 50, 80, etc.), or the first average intrusion distance is greater than or equal to the ninth preset threshold (flexibly set, such as 30 pixels, 50 pixels, 200 pixels, etc.), it is determined that the parking space frame is severely occupied and cannot be adjusted / parked.

[0114] In this embodiment, the obstacle data also includes a contour boundary line of the obstacle;

[0115] When the intrusion state is the first intrusion state, adjusting the space parking space frame according to the intrusion state and the obstacle data to obtain a target space parking space frame for guiding the vehicle to park may further include:

[0116] When the intrusion state is the first intrusion state, determining, in the contour boundary line, a line segment having a distance from the space parking space frame less than a fourth preset threshold, an angle less than a preset angle and the longest length as a reference boundary line;

[0117] Adjusting the direction of the space parking space frame so that the space parking space frame is parallel to the reference boundary line;

[0118] Adjusting the space parking space frame according to the reference boundary line to obtain an adjusted space parking space frame;

[0119] The adjusted space parking space frame is adjusted according to other line segments in the contour boundary line except the reference boundary line to obtain the target space parking space frame.

[0120] In this embodiment, from the contour boundary lines detected by the vehicle, a line segment having a distance from the spatial parking space frame less than a fourth preset threshold, an angle less than a preset angle and the longest length is selected as a reference boundary line.

[0121] It is understandable that the contour boundary line detected in this embodiment represents the contour boundary of the obstacle, which is composed of multiple line segments. In this embodiment, the distance between each line segment in the contour boundary line and the spatial parking space frame can be calculated by calculating the distance between each line segment and the four sides of the spatial parking space frame, and taking the minimum value as the calculation result. Among them, if the endpoints and the center point coordinates of each line segment of the contour boundary line are known, the straight line expression of the line segment can be calculated, and then the distance between the line segment and the sides of the spatial parking space frame is calculated based on the straight line expression. The specific calculation process is a conventional technical means for calculating the straight line distance, which will not be repeated here.

[0122] In this embodiment, the fourth preset threshold, the fifth preset threshold and the sixth preset threshold can be flexibly set according to user needs, such as 100 pixels, 200 pixels, 500 pixels, etc. The preset angle can be flexibly set according to user needs, such as 10°, 30°, etc. The angle between the line segment and the spatial parking space frame can be selected as the reference edge of the spatial parking space frame closest to the line segment, and the angle between the line segment and the reference edge is calculated as the angle between the line segment and the spatial parking space frame.

[0123] In this embodiment, after determining the reference boundary line, the direction of the spatial parking space frame is adjusted to be parallel to the reference boundary line. The direction vector of the reference boundary line can be calculated based on the coordinates of the two end points and the center point of the reference boundary line, and any side edge of the spatial parking space frame is used as a reference edge to adjust the direction of the spatial parking space frame so that the direction vector of the reference edge of the spatial parking space frame is parallel to the direction vector of the reference boundary line.

[0124] In this embodiment, adjusting the space parking space frame according to the reference boundary line to obtain the adjusted space parking space frame may include:

[0125] When there are other vehicles in front of the spatial parking space frame, and the minimum distance between the circumscribed rectangular placeholder frame corresponding to the other vehicles and the reference boundary line is less than a fifth preset threshold, the direction of the spatial parking space frame is adjusted so that the distance between the side of the spatial parking space frame closest to the reference boundary line and the reference boundary line is less than a sixth preset threshold, to obtain an adjusted spatial parking space frame.

[0126] Alternatively, the adjusting the space parking space frame according to the reference boundary line to obtain the adjusted space parking space frame may further include:

[0127] When there are other vehicles in front of the spatial parking space frame, and the minimum distance between the circumscribed rectangular placeholder frame corresponding to the other vehicles and the reference boundary line is greater than or equal to a fifth preset threshold, the position of the spatial parking space frame is adjusted so that a line connecting the center point of the circumscribed rectangular placeholder frame corresponding to the other vehicles and the center point of the spatial parking space frame is parallel to the reference boundary line, thereby obtaining the adjusted spatial parking space frame.

[0128] In this embodiment, after the spatial parking space frame is adjusted to be parallel to the reference boundary line, if there are other vehicles in front of the spatial parking space frame except the vehicle itself, the circumscribed rectangular frame corresponding to the vehicle in front of the spatial parking space frame in the obstacle data is called. Then, the spatial parking space frame is adjusted according to the minimum distance between the circumscribed rectangular frame corresponding to the vehicle in front of the spatial parking space frame and the reference boundary line. The minimum distance between the circumscribed rectangular frame corresponding to the vehicle in front of the spatial parking space frame and the reference boundary line can be calculated by calculating the distance from the coordinates of each corner point of the circumscribed rectangular frame corresponding to the vehicle in front of the spatial parking space frame to the reference boundary line, and taking the minimum value as the minimum distance between the circumscribed rectangular frame corresponding to the vehicle in front of the spatial parking space frame and the reference boundary line.

[0129] It is understandable that if there is no other vehicle in front of the vehicle, the subsequent correction line segment set construction and the spatial parking space frame adjustment can be directly performed.

[0130] In this embodiment, adjusting the adjusted space parking space frame according to other line segments in the contour boundary line except the reference boundary line to obtain the target space parking space frame may include:

[0131] The other line segments in the contour boundary line except the reference boundary line are taken as a modified line segment set, and the intrusion direction and intrusion distance corresponding to the third target pixel point in the modified line segment set are determined according to the coordinates corresponding to the third target pixel point in the modified line segment set, wherein the third target pixel point represents the endpoint and the midpoint of each line segment in the modified line segment set located in the adjusted spatial parking space frame;

[0132] Taking the set of the third target pixel points with the same invasion direction as a second unit set, and for each second unit set, determining a second average invasion distance of all the third target pixel points in the second unit set;

[0133] According to the invasion direction corresponding to all the third target pixel points in the second unit set and the second average invasion distance, the adjusted spatial parking space frame is adjusted to obtain the target spatial parking space frame, wherein the adjustment direction of the adjusted spatial parking space frame is the opposite direction of the invasion direction corresponding to all the pixel points in the second unit set, and the adjustment distance of the adjusted spatial parking space frame is the second average invasion distance plus the second preset tolerance gap.

[0134] In this embodiment, refer to Figure 5 a. After obtaining the adjusted parking space frame, the other line segments in the contour boundary line except the reference boundary line are used as the correction line segment set, and the endpoints and midpoints of each line segment in the correction line segment set located in the adjusted parking space frame are used as the third target pixel point. Then, according to the coordinates of the third target pixel point, the intrusion direction and intrusion distance corresponding to each third target pixel point are determined. Figure 5 In a, the endpoint of the lower end of the line segment located in the space parking space frame is the third target pixel point.

[0135] In this embodiment, the specific method of adjusting the adjusted space parking space frame by the second unit set to obtain the target space parking space frame can refer to the above method of adjusting the space parking space frame by the first unit set. The difference between the two is only that the pixels constituting the unit set are different, which will not be described here. Figure 5 b and Figure 6 , Figure 5 b is an example of the result after adjusting the adjusted space parking space frame through the second unit set. Figure 6 That is, it is a bird's-eye view after the adjusted space parking space frame is adjusted by the second unit set.

[0136] It is understandable that in actual applications, when the intrusion state of the spatial parking space frame is in an unadjustable / unparkable state, the spatial parking space frame is directly abandoned and a new spatial parking space frame is found and generated. In actual applications, during the parking process, the target spatial parking space frame obtained by the initial adjustment can be used as a new spatial parking space frame, and steps 210 to 230 are executed repeatedly until the vehicle completes parking. In this way, the spatial parking space frame can be adjusted in real time according to the obstacles around the vehicle, avoiding the appearance of new obstacles in the spatial parking space frame due to the incomplete receptive field of the on-board sensor, resulting in accidents such as vehicle collisions during parking.

[0137] For easier understanding, please refer to Figure 7 , the overall implementation process of the parking space adjustment method will be described in detail below:

[0138] s1: Data acquisition and preprocessing, for the input space parking space, laser fs (freespace point, indicating the point of the object detected by the laser, i.e. laser point cloud), laser OD (Object Detection: the object target recognition result detected by the laser point cloud, which is a three-dimensional box or a two-dimensional box, i.e. the circumscribed rectangular placeholder box), surround view fs (freespace point detected by the surround view fisheye, indicating the point of the object perceived by the surround view, i.e. the surround view point cloud) and ultrasonic cluster (clustering result of ultrasonic detection of obstacle boundaries, in the form of line segments, i.e. contour boundary lines) are time synchronized. The three-dimensional laser OD is compressed to a two-dimensional plane and converted to a two-dimensional OD. For the ultrasonic cluster, it is converted to the local data format (including the coordinates of the two endpoints and the center point).

[0139] s2: OD occupancy detection, for the input parking space, traverse all laser ODs and calculate the IOU area with the parking space. If the IOU area exceeds the threshold, the parking space is considered to be occupied by OD and the parking space is set to occupied state.

[0140] S3: FS occupancy detection, for parking spaces not occupied by OD, traverse all laser FSs and find the set of FS points within the parking space frame. If the number of intrusions of laser FSs is greater than the threshold, the parking space is considered to be occupied by laser FSs.

[0141] s4: Traverse the look-around fs whose category attribute is curb, and find the look-around fs set located in the parking space. If the number of look-around fs intrusions exceeds the threshold, continue to calculate the ratio of the maximum eigenvalue to the second largest eigenvalue of the intruding look-around fs point, if it conforms to the straight line distribution characteristics. Specifically, calculate the covariance matrix of all intrusion point coordinates, and then perform eigenvalue decomposition on the covariance matrix to obtain the maximum eigenvalue and the second largest eigenvalue, and compare the two. If the maximum eigenvalue is much larger than the second largest eigenvalue (for example, 5 times), it is considered that fs is more in line with the straight line distribution. At this time, the look-around fs is a valid point, and the parking space is set to occupied.

[0142] s5: For parking spaces occupied by fs, calculate the intrusion distance from the intrusion point fs to the front and rear edges of the parking space. Points with intrusion distances within a certain range are counted as adjustment reference points, and the rest are points that cannot be adjusted. When adjusting, the adjustment reference points are divided into two categories: upper intrusion point set and lower intrusion point set, and the classification is based on which side has the shortest distance to the upper and lower edges. Check the number of points that cannot be adjusted. If the number exceeds the threshold, the parking space is considered to be severely occupied and no adjustment is performed.

[0143] s6: For parking spaces that are not severely occupied, select the set with the largest number of points in the upper and lower intrusion point sets, and calculate the average intrusion distance of the intrusion points in the set. If the intrusion distance is greater than the threshold, the parking space is also considered to be severely occupied and the adjustment fails. Otherwise, the parking space is adjusted longitudinally. The adjustment method is: adjustment direction * adjustment distance, adjustment direction = the opposite direction of the intrusion direction, adjustment distance = average intrusion distance + tolerance gap.

[0144] s7: lateral adjustment of the parking space fs, with the left and right sides of the parking space as the basis for calculating the intrusion distance. Perform the same logical operation as s5 and s6 to perform lateral adjustment of the parking space.

[0145] s8: Reference cluster generation: In the stage of detecting parking spaces, for horizontal parking spaces (parking spaces whose long sides are consistent with the direction of the vehicle, generally parked in a side parking mode), find an ultrasonic cluster that is in the same longitudinal range as the parking space and close to the parking space, calculate the angle between the cluster and the parking space direction, and if the angle is within the threshold range and the cluster length exceeds the threshold, add it to the candidate reference cluster. Finally, select the longest cluster among the candidate clusters as the reference cluster.

[0146] s9: Based on the direction adjustment of the parking space of the reference cluster, the direction of the reference cluster is calculated and the parking space is adjusted to be parallel to the reference cluster.

[0147] s10: Parking space position adjustment based on the front vehicle OD and reference cluster: Calculate the shortest distance from the laser OD frame of the front vehicle to the reference cluster direction. If the distance is within the threshold range, adjust the parking space to the curb cluster. If the distance is greater than the threshold, align the center of the parking space with the center of the front vehicle laser OD in the reference cluster direction.

[0148] s11: Generation of adjusted cluster points: In the parking phase, for all parking space types, all clusters are traversed. The geometric relationship between the two endpoints and the center point and the parking space frame is calculated. If the point is located in the parking space, the point is added to the set of adjusted cluster points.

[0149] s12: Parking space adjustment based on cluster points: The invaded cluster points are processed according to the same logic as the fs points, using the adjustment method of the fs points in s5-s7, but ignoring other restrictions such as the number of invasion points to adjust the parking space and determine the occupancy status at the same time.

[0150] S13: Parking space availability judgment, finally, for the parking spaces that have been successfully adjusted by FS or the parking spaces that have been successfully adjusted by ultrasonic cluster, the occupancy attribute is set to unoccupied. For the parking spaces that are occupied by OD or failed to be adjusted by FS, the occupancy attribute is set to occupied.

[0151] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the electronic device 100 described above can refer to the corresponding process of each step of the aforementioned method, and will not be elaborated herein.

[0152] The embodiment of the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed on a computer, the computer executes the parking space adjustment method described in the above embodiment.

[0153] The present application also provides a computer program product, including a computer program. When the computer program is executed by the processor 101, the parking space adjustment method described in the above embodiment can be implemented.

[0154] Through the description of the above implementation methods, technical personnel in this field can clearly understand that the present application can be implemented by hardware, and can also be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0155] In summary, the embodiments of the present application provide a parking space adjustment method, an electronic device, a storage medium and a program product. In the technical solution, first, a space parking space frame and obstacle data generated when a vehicle is in a parking environment without parking space lines are obtained. Then, according to the space parking space frame and the obstacle data, based on a preset strategy, the intrusion state of the obstacle to the space parking space frame is determined. Finally, according to the intrusion state and the obstacle data, the space parking space frame is adjusted to obtain a target space parking space frame for guiding the parking of the vehicle. In this way, a space parking space is generated based on an environment without parking space lines, and the intrusion state of the obstacles around the vehicle to the space parking space is judged based on the obstacle data, and then the space parking space is adjusted based on the intrusion state and the obstacle data, so as to avoid the generated space parking space being limited by the receptive field of the vehicle sensor, so that new obstacles appear in the space parking space during the parking process of the vehicle, and there is a risk of collision between the vehicles. Although the traditional parking space division method can generate space parking spaces not based on parking lines, it is not flexible enough and cannot flexibly adjust the generated parking space frame.

[0156] In the embodiments provided in the present application, it should be understood that the disclosed device, system and method can also be implemented in other ways. The above-described device, system and method embodiments are merely schematic, for example, the flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a part of a module, a program segment or a code, and a part of the module, program segment or code contains one or more executable instructions for implementing the specified logical function. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. In addition, each functional module in each embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.

[0157] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A parking space adjustment method, characterized in that: The method comprises: Obtain the spatial parking space frame and obstacle data generated when the vehicle is in a parking environment without parking space lines; According to the space parking space frame and the obstacle data, based on a preset strategy, determining the intrusion state of the obstacle into the space parking space frame; The spatial parking space frame is adjusted according to the intrusion state and the obstacle data to obtain a target spatial parking space frame for guiding the parking of the vehicle.

2. The method according to claim 1, characterized in that The obstacle data includes a circumscribed rectangular placeholder frame and point cloud data of the obstacle; The determining, based on the spatial parking space frame and the obstacle data and a preset strategy, of an intrusion state of an obstacle into the spatial parking space frame includes: Determining an intersection-and-joint ratio of the spatial parking space frame to the circumscribed rectangular placeholder frame according to the spatial parking space frame and the circumscribed rectangular placeholder frame; When the intersection-over-union ratio is less than a first preset threshold, the intrusion state is determined according to the spatial parking space frame and the point cloud data.

3. The method according to claim 2, characterized in that The point cloud data includes a laser point cloud mapped to the same plane as the parking space frame and first two-dimensional coordinates corresponding to pixel points in the laser point cloud; When the intersection-over-union ratio is less than a first preset threshold, determining the intrusion state according to the spatial parking space frame and the point cloud data includes: When the intersection-over-union ratio is less than the first preset threshold, determining the number of first target pixel points according to the spatial parking space frame and the first two-dimensional coordinates, where the first target pixel points represent the pixel points whose first two-dimensional coordinates are located in the spatial parking space frame; When the number of the first target pixel points is greater than or equal to a second preset threshold, it is determined that the intrusion state is a first intrusion state indicating that an obstacle is contained in the space parking space frame.

4. The method according to claim 2, characterized in that: The point cloud data includes a second two-dimensional coordinate corresponding to a surround view point cloud and a pixel point in the surround view point cloud mapped to the same plane as the parking space frame; When the intersection-over-union ratio is less than a first preset threshold, determining the intrusion state according to the spatial parking space frame and the point cloud data includes: When the intersection-over-union ratio is less than the first preset threshold, determining the number of second target pixel points according to the spatial parking space frame and the second two-dimensional coordinates, where the second target pixel points represent the pixel points whose second two-dimensional coordinates are located in the spatial parking space frame; When the number of the second target pixel points is greater than or equal to a third preset threshold, determining a covariance matrix corresponding to the second target pixel points according to the coordinates of the second target pixel points; Performing eigenvalue decomposition on the covariance matrix to obtain the largest eigenvalue and the second largest eigenvalue; When the magnitudes of the maximum eigenvalue and the second largest eigenvalue meet a preset condition, it is determined that the intrusion state is a first intrusion state indicating that an obstacle is contained in the space parking space frame.

5. The method according to claim 3 or 4, characterized in that: When the intrusion state is the first intrusion state, adjusting the space parking space frame according to the intrusion state and the obstacle data to obtain a target space parking space frame for guiding the parking of the vehicle includes: When the intrusion state is the first intrusion state, determining the intrusion direction and intrusion distance corresponding to each pixel point according to the coordinates corresponding to each pixel point in the point cloud data; Taking the set of the pixel points with the same invasion direction as a first unit set, and for each first unit set, determining a first average invasion distance of all the pixel points in the first unit set; According to the intrusion direction corresponding to all the pixel points in the first unit set and the first average intrusion distance, the spatial parking space frame is adjusted to obtain the target spatial parking space frame, wherein the adjustment direction of the spatial parking space frame is the opposite direction of the intrusion direction corresponding to all the pixel points in the first unit set, and the adjustment distance of the spatial parking space frame is the first average intrusion distance plus a preset tolerance gap.

6. The method according to claim 3 or 4, characterized in that: The obstacle data also includes a contour boundary line of the obstacle; When the intrusion state is the first intrusion state, adjusting the space parking space frame according to the intrusion state and the obstacle data to obtain a target space parking space frame for guiding the parking of the vehicle includes: When the intrusion state is the first intrusion state, determining, in the contour boundary line, a line segment having a distance from the space parking space frame less than a fourth preset threshold, an angle less than a preset angle and the longest length as a reference boundary line; Adjusting the direction of the space parking space frame so that the space parking space frame is parallel to the reference boundary line; Adjusting the space parking space frame according to the reference boundary line to obtain an adjusted space parking space frame; The adjusted space parking space frame is adjusted according to other line segments in the contour boundary line except the reference boundary line to obtain the target space parking space frame.

7. The method according to claim 6, characterized in that The step of adjusting the space parking space frame according to the reference boundary line to obtain an adjusted space parking space frame includes: When there are other vehicles in front of the spatial parking space frame, and the minimum distance between the circumscribed rectangular placeholder frame corresponding to the other vehicles and the reference boundary line is less than a fifth preset threshold, the direction of the spatial parking space frame is adjusted so that the distance between the side of the spatial parking space frame closest to the reference boundary line and the reference boundary line is less than a sixth preset threshold, to obtain the adjusted spatial parking space frame.

8. The method according to claim 6, characterized in that The step of adjusting the space parking space frame according to the reference boundary line to obtain an adjusted space parking space frame includes: When there are other vehicles in front of the spatial parking space frame, and the minimum distance between the circumscribed rectangular placeholder frame corresponding to the other vehicles and the reference boundary line is greater than or equal to a fifth preset threshold, the position of the spatial parking space frame is adjusted so that a line connecting the center point of the circumscribed rectangular placeholder frame corresponding to the other vehicles and the center point of the spatial parking space frame is parallel to the reference boundary line, thereby obtaining the adjusted spatial parking space frame.

9. The method according to claim 6, characterized in that The adjusting the adjusted space parking space frame according to other line segments in the contour boundary line except the reference boundary line to obtain the target space parking space frame includes: The other line segments in the contour boundary line except the reference boundary line are taken as a modified line segment set, and the intrusion direction and intrusion distance corresponding to the third target pixel point in the modified line segment set are determined according to the coordinates corresponding to the third target pixel point in the modified line segment set, wherein the third target pixel point represents the endpoint and the midpoint of each line segment in the modified line segment set located in the adjusted spatial parking space frame; Taking the set of the third target pixel points with the same invasion direction as a second unit set, and for each second unit set, determining a second average invasion distance of all the third target pixel points in the second unit set; According to the invasion direction corresponding to all the third target pixel points in the second unit set and the second average invasion distance, the adjusted spatial parking space frame is adjusted to obtain the target spatial parking space frame, wherein the adjustment direction of the adjusted spatial parking space frame is the opposite direction of the invasion direction corresponding to all the pixel points in the second unit set, and the adjustment distance of the adjusted spatial parking space frame is the second average invasion distance plus the second preset tolerance gap.

10. The method according to claim 1, characterized in that Between the step of obtaining the space parking space frame and obstacle data generated when the vehicle is in a parking environment without parking space lines, and determining the intrusion state of the obstacle into the space parking space frame based on a preset strategy according to the space parking space frame and the obstacle data, the method further includes: Preprocessing the obstacle data to perform time synchronization and format conversion on the obstacle data to obtain preprocessed obstacle data; The determining, based on the spatial parking space frame and the obstacle data and a preset strategy, of an intrusion state of an obstacle into the spatial parking space frame includes: According to the spatial parking space frame and the pre-processed obstacle data, based on a preset strategy, determining an intrusion state of the obstacle into the spatial parking space frame; The step of adjusting the space parking space frame according to the intrusion state and the obstacle data to obtain a target space parking space frame for guiding the vehicle to park includes: The spatial parking space frame is adjusted according to the intrusion state and the pre-processed obstacle data to obtain a target spatial parking space frame for guiding the parking of the vehicle.

11. An electronic device, characterized in that: The electronic device comprises a processor and a memory coupled to each other, wherein the memory stores a computer program. When the computer program is executed by the processor, the electronic device executes the method according to any one of claims 1 to 10.

12. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 10.

13. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 10.

Images