Parking method of paving robot, storage medium and paving robot

Abstract

The invention discloses a parking method of a paving robot, a storage medium and the paving robot, and the parking method comprises the steps: collecting the image information of a target area, and determining the coordinate origin of the installation of a photovoltaic module according to the image information; according to the coordinate origin and the installation size of the photovoltaic module to be installed, the installation coordinate of each photovoltaic module on the photovoltaic support is determined; and the parking position of the paving robot is determined according to the installation coordinates and the reachable range of a mechanical arm of the paving robot. By means of the parking method, the paving robot can achieve automatic parking, manual operation is omitted, and the parking efficiency and the parking precision of the paving robot are improved.

Description

Technical Field

[0001] This application relates to the field of photovoltaic installation technology, specifically to a parking method for a paving robot, a storage medium, and a paving robot. Background Technology

[0002] Photovoltaic module installation robots have been gradually applied to the construction of centralized photovoltaic power plants, which has improved installation efficiency to a certain extent, reduced the labor intensity of workers, and reduced the demand for manpower.

[0003] During photovoltaic (PV) installation, the installation robot needs to be precisely positioned according to different installation requirements. However, in related technologies, the robot's positioning is usually done manually. This method is not only inefficient but also susceptible to variations in operator skill and experience, resulting in inconsistent positioning performance between different operators and affecting the robot's installation effect on the PV modules. Summary of the Invention

[0004] The purpose of this application is to provide a parking method, storage medium, and parking robot for a paving robot, aiming to solve the problems of low efficiency and poor accuracy of manual parking in related technologies.

[0005] To achieve the objectives of this application, in a first aspect, this application provides a parking method for a paving robot, comprising:

[0006] Acquire image information of the target area and determine the coordinate origin of the photovoltaic module installation based on the image information;

[0007] Based on the coordinate origin and the installation dimensions of the photovoltaic modules to be installed, determine the installation coordinates of each photovoltaic module on the photovoltaic bracket;

[0008] The parking position of the paving robot is determined based on the installation coordinates and the reachability of the robotic arm of the paving robot.

[0009] In one possible implementation, acquiring image information of the target area and determining the coordinate origin of the photovoltaic module based on the image information includes:

[0010] If there are photovoltaic modules in the target area, one of the photovoltaic modules is selected as the target photovoltaic module, and the coordinate origin of the photovoltaic panel installation is determined based on the target photovoltaic module;

[0011] If there are no photovoltaic modules in the target area, the coordinate origin for the installation of the photovoltaic modules is determined based on the photovoltaic support structure.

[0012] In one possible implementation, the target photovoltaic module includes a first profile line and a second profile line, the first profile line and the second profile line being perpendicular to each other. Determining the coordinate origin for photovoltaic panel installation based on the target photovoltaic module includes the following steps:

[0013] Obtain the coordinates of the first key point of the first contour line and the coordinates of the second key point of the second contour line;

[0014] Fit the coordinates of the first key point to obtain the equation of the first straight line, and fit the coordinates of the second key point to obtain the equation of the second straight line;

[0015] Determine the coordinates of the first intersection point of the first straight line equation and the second straight line equation, and set the coordinates of the first intersection point as the origin of the coordinate system for the installation of the photovoltaic module.

[0016] In one possible implementation, the target photovoltaic module includes multiple solar cells, and obtaining the coordinates of a first key point of the first contour line and the coordinates of a second key point of the second contour line includes:

[0017] The positions of the endpoints of multiple battery cells, the first contour line, and the second contour line are determined using the image information.

[0018] Based on the position of the endpoint of the battery cell and the first contour line, determine the coordinates of the second intersection point of the endpoint of the battery cell and the first contour line, and set the coordinates of the second intersection point as the coordinates of the first key point;

[0019] Based on the position of the endpoint of the battery cell and the second contour line, determine the coordinates of the third intersection point of the endpoint of the battery cell and the second contour line, and set the coordinates of the third intersection point as the coordinates of the second key point.

[0020] In one possible implementation, the photovoltaic support includes a third profile line and a fourth profile line that are perpendicular to each other, and the determination of the coordinate origin for photovoltaic module installation based on the photovoltaic support includes:

[0021] Obtain the coordinates of the third key point of the third contour line and the coordinates of the fourth key point of the fourth contour line;

[0022] Fit the coordinates of the third key point to obtain the equation of the third straight line, and fit the coordinates of the fourth key point to obtain the equation of the fourth straight line;

[0023] Determine the coordinates of the fourth intersection point of the equations of the third and fourth lines, and set the coordinates of the fourth intersection point as the origin of the coordinate system for the installation of the photovoltaic module.

[0024] In one possible implementation, the photovoltaic support includes a first support and a second support perpendicular to each other, with the first outline and the second outline disposed on the first support; or

[0025] The first contour line and the second contour line are disposed on the second bracket; or

[0026] The first outline is located on the first bracket, and the second outline is located on the second bracket.

[0027] In one possible implementation, determining the parking position of the paving robot based on the installation coordinates and the reachability of the paving robot's robotic arm includes:

[0028] The reachable range of the robotic arm is projected onto the mounting plane of the photovoltaic module to obtain the movement trajectory of the robotic arm on the mounting plane;

[0029] The parking position of the paving robot is determined based on the motion trajectory and the installation coordinates of each photovoltaic module.

[0030] In one possible implementation, determining the parking position of the paving robot based on the motion trajectory and the installation coordinates of each photovoltaic module includes:

[0031] Calculate the maximum area that the motion trajectory can cover for each of the installation coordinates, and obtain the target coordinates of the motion trajectory when the coverage area of ​​the motion trajectory on the installation coordinates reaches the maximum.

[0032] Calculate the relative position of the paving robot and the photovoltaic support when the motion trajectory reaches the target coordinates;

[0033] The parking position of the paving robot is determined based on the relative position and environmental factors.

[0034] Secondly, this application also proposes a storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, cause the processor to execute a parking method for a paving robot, the parking method for the paving robot including:

[0035] Acquire image information of the target area and determine the coordinate origin of the photovoltaic module installation based on the image information;

[0036] Based on the coordinate origin and the installation dimensions of the photovoltaic modules to be installed, determine the installation coordinates of each photovoltaic module on the photovoltaic bracket;

[0037] The parking position of the paving robot is determined based on the installation coordinates and the reachability of the robotic arm of the paving robot.

[0038] Thirdly, this application also proposes a paving robot for paving photovoltaic modules. The paving robot includes a processor, a memory, and a robotic arm. The memory stores computer-readable instructions, and the processor invokes the instructions stored in the memory to execute a parking method for the paving robot. The parking method for the paving robot includes:

[0039] Acquire image information of the target area and determine the coordinate origin of the photovoltaic module installation based on the image information;

[0040] Based on the coordinate origin and the installation dimensions of the photovoltaic modules to be installed, determine the installation coordinates of each photovoltaic module on the photovoltaic bracket;

[0041] The parking position of the paving robot is determined based on the installation coordinates and the reachability of the robotic arm of the paving robot.

[0042] This application proposes a parking method applicable to paving robots. This method allows the paving robot to first determine the coordinate origin for photovoltaic module installation by acquiring image information of the target area. Then, using the coordinate origin and the installation dimensions of the photovoltaic modules to be installed, the installation coordinates of each photovoltaic module on the photovoltaic support are determined. Finally, based on the installation coordinates and the reachability of the paving robot's robotic arm, the parking position of the paving robot is determined, thereby achieving automatic parking of the paving robot, eliminating manual operation, and improving the parking efficiency and accuracy of the paving robot. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0044] Figure 1 A schematic diagram of the operating environment of the paving robot provided in this application;

[0045] Figure 2 A flowchart illustrating the parking method for the paving robot provided in this application;

[0046] Figure 3 This is a mask image of the photovoltaic module described in this application;

[0047] Figure 4 This is a schematic diagram of the structure of the photovoltaic module in this application;

[0048] Figure 5A detailed schematic diagram of the first outline of the photovoltaic module of this application;

[0049] Figure 6 This is a schematic diagram of the photovoltaic support structure of this application;

[0050] Figure 7 for Figure 6 Detailed schematic diagram of the photovoltaic support structure;

[0051] Figure 8 This is a structural diagram of the third and fourth contour lines of this application in the region near their upper endpoints;

[0052] Figure 9 This is a structural diagram of the third and fourth contour lines of this application in the region near their lower endpoints;

[0053] Figure 10 This is a structural diagram of the third and fourth contour lines of this application in the region near the left endpoint;

[0054] Figure 11 This is a schematic diagram of the structure of the third and fourth contour lines of this application in the region near the right endpoint;

[0055] Figure 12 This is a structural schematic diagram of the third outline of this application in the first bracket;

[0056] Figure 13 This is a schematic diagram of the structure of the fourth outline of this application when it is in the second bracket.

[0057] Explanation of reference numerals in the attached figures:

[0058] 1-Photovoltaic module, 11-Solar cell, 11a-First row of solar cells, 11b-First column of solar cells, 111-End point, 111a-First end point, 111b-Second end point, 12-First outline, 121-Rectangular region, 122-Visible end point, 123-Invisible end point, 124-First straight line equation, 13-Second outline;

[0059] 1' - Background;

[0060] 2-Photovoltaic support, 2a-Target support, 21-First support, 211-Area near the upper end point, 212-Area near the lower end point, 22-Second support, 221-Area near the left end point, 222-Area near the right end point, 23-Installation area, 24-Third outline, 25-Fourth outline, 26-Equation of the third straight line, 27-Equation of the fourth straight line;

[0061] 241 - Third key point; 251 - Fourth key point. Detailed Implementation

[0062] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0063] It should be noted that when a component is said to be "fixed" to another component, it can be directly on the other component or it can be in a middle component. When a component is said to be "connected" to another component, it can be directly connected to the other component or it may be in a middle component.

[0064] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The term "and / or" as used in this application includes any and all combinations of one or more of the associated listed items.

[0065] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0066] This application proposes a paving robot for installing photovoltaic (PV) modules onto PV mounting brackets. The paving robot includes a processor, a memory, a 3D camera, a moving mechanism, and a robotic arm mounted on the moving mechanism. The robotic arm is used to grasp or place the PV modules, thereby achieving the installation of the PV modules on the PV mounting brackets. The moving mechanism drives the robotic arm to move relative to the PV mounting brackets, allowing the robotic arm to flexibly move between different installation positions to adapt to the installation requirements of PV modules in different locations.

[0067] A 3D camera is used to acquire images of the target area for photovoltaic module installation, thereby assisting the installation robot in completing the installation of the photovoltaic modules. In one embodiment, the 3D camera can be mounted on a robotic arm and located at the front end of the robotic arm. After the robotic arm grasps the photovoltaic module and moves it near the photovoltaic support, the 3D camera is used to acquire image information of the photovoltaic support, thereby assisting the installation robot in completing the installation of the photovoltaic module.

[0068] In other possible implementations, the 3D camera can also be separate from the robotic arm. The photovoltaic module installation equipment includes a drone module, which includes a 3D camera. After the robotic arm grasps the photovoltaic module and moves it near the photovoltaic support, the drone module flies above the photovoltaic support and the 3D camera mounted thereon takes a picture. Alternatively, before the robotic arm grasps the photovoltaic module and moves it near the photovoltaic support, the drone module flies above the photovoltaic support and takes a picture in advance.

[0069] Please refer to Figure 1 The memory 1005 is used to store computer instructions, and is used to store the operating system, network communication module, user interface module, and control program for the laying robot. The memory 1005 can be volatile memory or non-volatile memory, or may include both. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).

[0070] The processor 1001 is used to invoke the computer instructions to execute the parking method of the paving robot. The processor 1001 may be a central processing unit (CPU), or it may 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 gate or transistor logic devices, discrete hardware components, etc.

[0071] In other possible implementations, the paving robot also includes a communication bus 1002, a user interface 1003, a network interface 1004, and a memory. The communication bus 1002 is used to enable communication between these components. The user interface 1003 is mainly used for user data interaction and may include a display screen, an input unit such as a keyboard, and optionally, a standard wired interface or a wireless interface. The network interface 1004 is mainly used for data communication with a network server and may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface).

[0072] During photovoltaic (PV) installation, the installation robot needs to be precisely positioned according to different installation requirements. However, in related technologies, the robot's positioning is usually done manually. This method is not only inefficient but also susceptible to variations in operator skill and experience, resulting in inconsistent positioning performance between different operators and affecting the robot's installation effect on the PV modules.

[0073] To address the aforementioned issues, this application also proposes a storage medium and a parking method for a paving robot. The storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions cause the processor to execute the parking method for the paving robot.

[0074] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the parking method for the paving robot provided in this application. The parking method for the paving robot includes:

[0075] S101. Collect image information of the target area and determine the coordinate origin of the photovoltaic module installation based on the image information.

[0076] In practical applications, the paving robot first uses a 3D camera to capture images of the target area where photovoltaic modules need to be installed, obtaining image information of the target area. This image information includes images of the photovoltaic support structure and the photovoltaic modules installed on the support structure. Alternatively, in some other possible implementations, there may not be photovoltaic modules installed on the photovoltaic support structure in the target area; in this case, the information captured by the 3D camera will only be the image information of the photovoltaic support structure.

[0077] It should be noted that the image information acquired by the 3D camera can be either 3D or planar image information, and this application does not impose any limitation on this. In one embodiment, the image acquired by the 3D camera is an RGB (Red, Green, Blue, RGB) image. In an RGB image, each pixel is represented by three values, corresponding to the three colors red, green, and blue, respectively. For example, a pixel can be represented as (R, G, B), where R, G, and B are all integers between 0 and 255. The processor converts the RGB image of the target area into a grayscale image, applies filtering to reduce noise, and then segments the image of the target area using the YOLO (You Only Look Once, YOLO) instance segmentation algorithm to obtain the image information of the photovoltaic support and photovoltaic module.

[0078] If the segmented image does not contain an image of the photovoltaic (PV) module, it indicates that PV module installation has not yet begun in that target area. In this case, the processor will select the PV support as a reference to determine the installation point of the PV module on the PV support, and use this as the origin of the coordinate system for PV module installation.

[0079] If there are already installed photovoltaic modules in the segmented image, the processor will use the installed photovoltaic modules as references to determine the installation points of subsequent photovoltaic modules, and use these as the coordinate origin for the installation of photovoltaic modules.

[0080] Specifically, in one embodiment of this application, acquiring image information of the target area and determining the coordinate origin of the photovoltaic module based on the image information includes:

[0081] If photovoltaic modules exist in the target area, one of the photovoltaic modules is selected as the target photovoltaic module, and the coordinate origin of the photovoltaic panel installation is determined based on the target photovoltaic module.

[0082] Specifically, after obtaining the image information of the target area, the processor will segment the image of the target area using the YOLO (You Only LookOnce) instance segmentation algorithm. If there is an image of a photovoltaic module in the segmented image, the processor will select one of the photovoltaic module images as the target photovoltaic module and use the target photovoltaic module as a reference to determine the coordinate origin of the photovoltaic module installation.

[0083] In practical applications, the processor selects an image of one of the photovoltaic modules as the target photovoltaic module and uses this target photovoltaic module as a reference to determine the coordinate origin of the photovoltaic module installation. The steps include the following:

[0084] (1) Label the bounding box and corresponding binary mask of each photovoltaic module within the target area. The bounding box of the photovoltaic module includes its width and height. Please refer to [link to relevant documentation]. Figure 3 A binary mask is a binary image where each pixel has a value of either 0 or 255. In image processing: 0 (black): represents background 1'; 255 (white): represents photovoltaic module 1. The binary mask is used to separate the photovoltaic module from the background.

[0085] (2) The RGB image of the target area is segmented according to the bounding box of each photovoltaic module and the corresponding binary mask in the marked target area to obtain the binary mask image of each photovoltaic module.

[0086] (3) Select one photovoltaic module as the target photovoltaic module. For ease of calculation, generally, the leftmost or rightmost photovoltaic module will be selected as the target photovoltaic module, depending on the installation direction of the photovoltaic modules. For example, if the installation of photovoltaic modules starts from the left side of the already installed photovoltaic modules, the photovoltaic module in the upper left or lower left corner of the already installed photovoltaic modules will be selected as the target photovoltaic module. If the installation of photovoltaic modules starts from the right side of the already installed photovoltaic modules, the photovoltaic module in the upper right or lower right corner of the already installed photovoltaic modules will be selected as the target photovoltaic module.

[0087] For clarity, we'll use the photovoltaic module in the bottom left corner as an example. To obtain this module, the processor first identifies the center point of each module. Assuming the image coordinate system has the bottom left corner as the origin, the X-axis to the right, and the Y-axis upward, the processor finds the center point with the smallest Y-coordinate, which is the lowest-positioned photovoltaic module. If multiple modules have the same smallest Y-coordinate, the module with the smallest X-coordinate, i.e., the leftmost module, is selected. This process is repeated to identify the photovoltaic module located in the bottom left corner.

[0088] (4) Read the binary mask area of ​​the target photovoltaic module, use the contour detection algorithm to detect all contours in the mask area, select the contour with the largest area and calculate the minimum bounding rectangle of the contour, and use this to obtain the edge contour of the target photovoltaic module in the image.

[0089] The edge profile includes a first profile line and a second profile line of the target photovoltaic panel near the photovoltaic module installation side. The first profile line and the second profile line are perpendicular to each other. That is, when the photovoltaic module is installed from the lower left corner, the first profile line and the second profile line are the left vertical edge and the lower horizontal edge of the target photovoltaic module, respectively.

[0090] Determining the coordinate origin for photovoltaic panel installation based on the first and second contour lines includes the following steps:

[0091] Obtain the coordinates of the first key point of the first contour line and the coordinates of the second key point of the second contour line.

[0092] Specifically, after obtaining the first and second contour lines, a point is selected as the origin with coordinates (0,0). A coordinate system is established by using the horizontal axis as the X-axis and the vertical axis as the Y-axis, starting from the origin. The coordinates (x, y) of the first key point are then obtained. i ,y i ) 1 , i = 1, ..., n, the coordinates of the second key point (x i ,y i ) 2 Let i = 1, ..., n. The YOLO keypoint detection algorithm is used to detect key points in the rectangular region 121, obtaining the first key point of the first contour line (horizontal side) and the second key point of the second contour line (vertical side) of the target photovoltaic module.

[0093] The key point detection algorithm has a good effect on handling key points that are not visible due to occlusion, high reflectivity, etc., and can supplement the accuracy of the instance segmentation algorithm. It ensures that the system can correctly and accurately find the edge of the target photovoltaic module when the instance segmentation algorithm fails to accurately segment the edge.

[0094] Fit the coordinates of the first key point to obtain the equation of the first straight line, and fit the coordinates of the second key point to obtain the equation of the second straight line.

[0095] After finding the first and second key points, the processor will use the least squares method to determine the coordinates (x, y) of the first and second key points. i ,y i ) 1 The first line equation of the first contour line and the second line equation of the second contour line are obtained by preprocessing the i = 1, ..., n and performing regression analysis.

[0096] Determine the coordinates of the first intersection point of the equations of the first and second straight lines, and set the coordinates of the first intersection point as the origin of the coordinate system for photovoltaic module installation.

[0097] Since the first and second contour lines are perpendicular to each other, the intersection of the first and second line equations is also unique. After obtaining the first and second line equations, the processor will obtain the intersection of the first and second line equations and the second line equations of the second contour lines, and use it as the coordinate origin for photovoltaic module installation.

[0098] Due to variations in lighting, object obstruction, or complex backgrounds, the initial outline of the photovoltaic module generated by the segmentation algorithm contains a certain range of errors, resulting in an incomplete initial outline of the photovoltaic module, which in turn leads to a large error between the first and second outlines.

[0099] To address the aforementioned issues, in one embodiment of this application, the target photovoltaic module includes multiple solar cells, each with four endpoints. Obtaining the coordinates of a first key point on a first contour line and the coordinates of a second key point on a second contour line includes:

[0100] The positions of the endpoints, first contour line, and second contour line of multiple battery cells are determined using image information.

[0101] Based on the position of the end point of the battery cell and the first contour line, determine the coordinates of the second intersection point of the end point of the battery cell and the first contour line, and set the coordinates of the second intersection point as the coordinates of the first key point.

[0102] Based on the positions of the endpoints of the battery cell and the second contour line, determine the coordinates of the third intersection point of the endpoints of the battery cell and the second contour line, and set the coordinates of the third intersection point as the coordinates of the second key point.

[0103] Please refer to Figure 4 Specifically, the photovoltaic module 1 is composed of multiple solar cells 11, each solar cell 11 including four endpoints 111. The processor selects the solar cell 11 closest to the first contour line 12 and uses it as the first row of solar cells 11a (when the first contour line 12 is the lower horizontal line of the photovoltaic module, the processor selects the bottommost row of solar cells as the first row of solar cells 11a). Then, the processor extracts the first endpoints 111a of each solar cell in the first row of solar cells 11a and compares the positions of these first endpoints 111a with the positions of the first contour line 12, thereby obtaining the first endpoints 111a that fall on the first contour line 12 and using them as the first key points.

[0104] The processor selects the cell closest to the second contour line 13 and uses it as the first column of cells 11b. (When the second contour line 13 is the left vertical line of the photovoltaic module, the processor selects the leftmost column of cells as the first column of cells 11b.) Then, the processor extracts the second endpoints 111b of each cell 11 in the first column of cells 11b that are close to the second contour line 13, compares the positions of these endpoints with the positions of the second contour line 13, and obtains the second endpoints 111b that fall on the second contour line 13, and uses them as the second key points.

[0105] Finally, the processor will use the least squares method to determine the coordinates (x, y) of the first and second keypoints. i ,y i ) 1 The first line equation of the first contour line and the second line equation of the second contour line are obtained by preprocessing the i = 1, ..., n and performing regression analysis.

[0106] Please refer to Figure 5 Taking the first contour line as an example, the first contour line generated by the segmentation algorithm can be regarded as a rectangular region 121 with a certain error. The endpoints of the battery cells near the first contour line are located within the rectangular region 121. Due to changes in lighting, object occlusion, or complex backgrounds, some endpoints 123 are invisible. The visible endpoints 122 within the rectangular region of the first contour line are detected by a key point detection algorithm, and the positions of the invisible endpoints 123 are predicted to supplement the endpoints of the battery cells within the rectangular region 121 of the first contour line. After supplementing the endpoints of the battery cells within the rectangular region 121 of the first contour line, the first straight line equation 124 of the first contour line is calculated based on the complete battery cell endpoints.

[0107] The coordinates of the visible endpoints of the battery cells located within the edge range of the first and second contour lines are detected by a key point detection algorithm, and the coordinates of the invisible endpoints of the battery cells are predicted. Then, a straight line is fitted based on the coordinates of the complete battery cell endpoints to obtain accurate first and second straight line equations.

[0108] If there are no photovoltaic modules in the target area, the coordinate origin for the installation of photovoltaic modules is determined based on the photovoltaic support structure.

[0109] For details, please refer to Figure 6The photovoltaic support 2 includes multiple vertically arranged first supports 21 and multiple horizontally arranged second supports 22. The first supports 21 and second supports 22 are perpendicular to each other. The processor selects the first support 21 or the second support 22 closest to the photovoltaic module installation area 23 as the target support 2a, and acquires the RGB image information of the target support using a 3D camera. The processor then uses the YOLO instance segmentation algorithm to segment the image of the target support 2a to obtain its binary mask. Next, the processor uses a contour detection algorithm to detect all contours in the mask area, selects the contour with the largest area, and calculates its minimum bounding rectangle. This is used to obtain the third and fourth contour lines of the target support 2a in the image. Based on the third and fourth contour lines, the coordinate origin for the photovoltaic module installation is determined using the photovoltaic support, including:

[0110] Obtain the coordinates of the third key point of the third contour line and the coordinates of the fourth key point of the fourth contour line;

[0111] Specifically, after obtaining the first and second contour lines, a point is selected as the origin with coordinates (0,0). A coordinate system is established by using the horizontal axis as the X-axis and the vertical axis as the Y-axis, starting from the origin. The coordinates of the third key point (x, y) are then obtained. i ,y i ) 1 , i = 1, ..., n, the coordinates of the fourth key point (x i ,y i ) 2 Let i = 1, ..., n. The YOLO keypoint detection algorithm is used to detect key points on the target support, obtaining the third key point of the third contour line (horizontal edge) and the fourth key point of the fourth contour line (vertical edge) of the target photovoltaic module.

[0112] The key point detection algorithm has a good effect on handling key points that are not visible due to occlusion, high reflectivity, etc., and can supplement the accuracy of the instance segmentation algorithm. It ensures that the system can correctly and accurately find the edge of the target photovoltaic module when the instance segmentation algorithm fails to accurately segment the edge.

[0113] Fit the coordinates of the third key point to obtain the equation of the third line, and fit the coordinates of the fourth key point to obtain the equation of the fourth line.

[0114] After finding the third and fourth key points, the processor will use the least squares method to determine the coordinates (x, y) of the third and fourth key points. i ,y i ) 1 The equations of the third line of the third contour line and the fourth line of the fourth contour line are obtained by preprocessing the i = 1, ..., n and performing regression analysis.

[0115] Determine the coordinates of the fourth intersection point of the equations of the third and fourth lines, and set the coordinates of the fourth intersection point as the origin of the coordinate system for photovoltaic module installation.

[0116] Since the first contour line and the second contour line are perpendicular to each other, the intersection point of the first straight line equation and the second straight line equation is also unique. After obtaining the first straight line equation of the first contour line and the second straight line equation of the second contour line, the processor will obtain the intersection point of the first straight line equation of the first contour line and the second straight line equation of the second contour line, and use it as the coordinate origin for photovoltaic module installation.

[0117] Please refer to the following for further explanation. Figure 7 The target support 2a can be the first support 21, the second support 22, or both. When the target support 2a is selected as the vertically positioned first support 21, the processor first uses the YOLO instance segmentation algorithm to annotate the binary mask image of the region 211 near the upper endpoint of the first support 21 or the binary mask image of the region 212 near the lower endpoint of the first support 21. Please refer to [reference needed]. Figure 8 and Figure 9 The processor separates the contour points of region 211 near the upper endpoint or region 212 near the lower endpoint of the first support 21 from the background to obtain preliminary third contour lines 24 and fourth contour lines 25. Finally, the processor extracts third key points 241 and fourth key points 251 near the third contour lines 24 and fourth contour lines 25 using a key point detection algorithm, and fits the third key points 241 and fourth key points 251 using the least squares method to form the third line equation 26 and the fourth line equation 27.

[0118] When the target support 2a is selected as the second support 22 arranged horizontally, the processor first uses the YOLO instance segmentation algorithm to annotate the binary mask image of the region 221 near the left end point of the second support 22 or the binary mask image of the region 222 near the right end point of the second support 22. Please refer to... Figure 10 and Figure 11 Next, the processor separates the contour points of region 221 near the left endpoint or region 222 near the right endpoint of the second support 22 from the background to obtain preliminary third contour lines 24 and fourth contour lines 25. Finally, the processor extracts third key points 241 and fourth key points 251 near the third contour lines 24 and fourth contour lines 25 using a key point detection algorithm, and fits the third key points 241 and fourth key points 251 using the least squares method to form the third line equation 26 and the fourth line equation 27.

[0119] Please refer to Figure 12 and Figure 13When the target support 2a selects the first support 21 and the second support 22, for the first support 21, the processor will use the YOLO instance segmentation algorithm to annotate the region 211 near the upper end point, the region 212 near the lower end point, and the binary mask image portion between the regions 211 and 212 near the upper and lower ends of the first support 21. Then, the processor will separate the regions 211 and 212 near the upper and lower ends of the first support 21 from the background, thereby obtaining a preliminary third contour line 24. Finally, the processor will use a keypoint detection algorithm to extract the third keypoints 241 of the regions 211 and 212 near the upper and lower ends, and fit the third keypoints 241 using the least squares method, thereby forming the third straight line equation 26. For the second support 22, the processor uses the YOLO instance segmentation algorithm to annotate the regions 221 near the left endpoint, 222 near the right endpoint, and the binary mask image portion between the two endpoints. Then, the processor separates these regions from the background to obtain a preliminary fourth contour line 25. Finally, the processor uses a keypoint detection algorithm to extract the fourth keypoints 251 from the regions 221 and 222 near the left and right endpoints, and fits the fourth keypoints 251 using the least squares method to form the fourth line equation 27.

[0120] S102. Determine the installation coordinates of each photovoltaic module on the photovoltaic support based on the coordinate origin and the installation dimensions of the photovoltaic modules to be installed.

[0121] Specifically, after obtaining the coordinate origin, the processor determines the installation coordinates of each photovoltaic module on the photovoltaic support based on the pre-input or real-time obtained photovoltaic module installation dimensions. For example, assuming the coordinate origin is (0, 0), a photovoltaic module has a length of 1 and a width of 1, and there is no installation gap between photovoltaic modules, then the installation coordinates of the four endpoints of the first photovoltaic module are (0, 0), (0, 1), (1, 0), and (1, 1). The coordinates of the four endpoints of the second photovoltaic panel to the right of the first photovoltaic panel are (1, 1), (1, 2), (1, 1), and (2, 2). The coordinates of the four endpoints of the third photovoltaic panel above the first photovoltaic panel are (0, 1), (1, 1), (0, 2), and (1, 2), and so on.

[0122] S103. Determine the parking position of the paving robot based on the installation coordinates and the reachability of the paving robot's robotic arm.

[0123] After obtaining the installation coordinates of each photovoltaic module, the processor will combine the reachability of the paving robot's robotic arm to determine the parking position of the paving robot, thereby ensuring that the paving robot can install the maximum number of photovoltaic modules in a single movement without changing the reachability of the robotic arm.

[0124] Specifically, in one embodiment of this application, determining the parking position of the paving robot based on its motion trajectory and the installation coordinates of each photovoltaic module includes:

[0125] The reachable range of the robotic arm is projected onto the mounting plane of the photovoltaic module to obtain the movement trajectory of the robotic arm on the mounting plane;

[0126] The parking position of the paving robot is determined based on its movement trajectory and the installation coordinates of each photovoltaic module.

[0127] In this embodiment, the processor projects the reachable range of the robotic arm onto the mounting plane of the photovoltaic modules, obtains the movement trajectory of the robotic arm under the mounting plane, and determines the parking position of the paving robot based on the movement trajectory and the mounting coordinates of each photovoltaic module. Specifically, in one embodiment of this application, determining the parking position of the paving robot based on the movement trajectory and the mounting coordinates of each photovoltaic module includes:

[0128] Calculate the maximum coverage area of ​​the motion trajectory for each installation coordinate, and obtain the target coordinates of the motion trajectory when the coverage area of ​​the motion trajectory for the installation coordinates reaches the maximum.

[0129] Calculate the relative position of the paving robot and the photovoltaic support when the trajectory reaches the target coordinates;

[0130] The parking location of the paving robot is determined based on its relative position and environmental factors.

[0131] In this embodiment, the processor overlaps the robotic arm's movement trajectory with the photovoltaic modules until the overlap area between the trajectory and the modules is maximized. Then, the processor obtains the coordinates of the trajectory on the mounting plane in this state and reverse-engineers the relative position of the paving robot and the photovoltaic support under these projected coordinates. Based on this relative position and the detection of environmental factors, the processor determines the paving robot's stopping position, thereby ensuring that the paving robot can lay the maximum number of photovoltaic modules in a single movement.

[0132] In the description of the embodiments of this application, it should be noted that the orientation or positional relationship of the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and other indicators are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0133] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Those skilled in the art will understand that all or part of the processes for implementing the above embodiments and equivalent variations made in accordance with the claims of this application are still within the scope of this application.

Claims

1. A parking method for a paving robot, characterized in that, include: Acquire image information of the target area and determine the coordinate origin of the photovoltaic module installation based on the image information; Based on the coordinate origin and the installation dimensions of the photovoltaic modules to be installed, determine the installation coordinates of each photovoltaic module on the photovoltaic support. The parking position of the paving robot is determined based on the installation coordinates and the reachability of the robotic arm of the paving robot.

2. The parking method for the paving robot as described in claim 1, characterized in that, The acquisition of image information of the target area and the determination of the coordinate origin of the photovoltaic module based on the image information include: If there are photovoltaic modules in the target area, one of the photovoltaic modules is selected as the target photovoltaic module, and the coordinate origin of the photovoltaic panel installation is determined based on the target photovoltaic module; If there are no photovoltaic modules in the target area, the coordinate origin for the installation of the photovoltaic modules is determined based on the photovoltaic support structure.

3. The parking method for the paving robot as described in claim 2, characterized in that, The target photovoltaic module includes a first contour line and a second contour line, the first contour line and the second contour line being perpendicular to each other. Determining the coordinate origin for photovoltaic panel installation based on the target photovoltaic module includes the following steps: Obtain the coordinates of the first key point of the first contour line and the coordinates of the second key point of the second contour line; Fit the coordinates of the first key point to obtain the equation of the first straight line, and fit the coordinates of the second key point to obtain the equation of the second straight line; Determine the coordinates of the first intersection point of the first straight line equation and the second straight line equation, and set the coordinates of the first intersection point as the origin of the coordinate system for the installation of the photovoltaic module.

4. The parking method for the paving robot as described in claim 3, characterized in that, The target photovoltaic module includes multiple solar cells, and obtaining the coordinates of the first key point of the first contour line and the coordinates of the second key point of the second contour line includes: The positions of the endpoints of multiple battery cells, the first contour line, and the second contour line are determined using the image information. Based on the position of the endpoint of the battery cell and the first contour line, determine the coordinates of the second intersection point of the endpoint of the battery cell and the first contour line, and set the coordinates of the second intersection point as the coordinates of the first key point; Based on the position of the endpoint of the battery cell and the second contour line, determine the coordinates of the third intersection point of the endpoint of the battery cell and the second contour line, and set the coordinates of the third intersection point as the coordinates of the second key point.

5. The parking method for the paving robot as described in claim 2, characterized in that, The photovoltaic support structure includes a third profile line and a fourth profile line that are perpendicular to each other. The determination of the coordinate origin for photovoltaic module installation based on the photovoltaic support structure includes: Obtain the coordinates of the third key point of the third contour line and the coordinates of the fourth key point of the fourth contour line; Fit the coordinates of the third key point to obtain the equation of the third straight line, and fit the coordinates of the fourth key point to obtain the equation of the fourth straight line; Determine the coordinates of the fourth intersection point of the equations of the third and fourth lines, and set the coordinates of the fourth intersection point as the origin of the coordinate system for the installation of the photovoltaic module.

6. The parking method for the paving robot as described in claim 3, characterized in that, The photovoltaic support includes a first support and a second support that are perpendicular to each other, with the first outline and the second outline disposed on the first support; or The first contour line and the second contour line are disposed on the second bracket; or The first outline is located on the first bracket, and the second outline is located on the second bracket.

7. The parking method for the paving robot as described in claim 6, characterized in that, Determining the parking position of the paving robot based on the installation coordinates and the reachability of the paving robot's robotic arm includes: The reachable range of the robotic arm is projected onto the mounting plane of the photovoltaic module to obtain the movement trajectory of the robotic arm on the mounting plane; The parking position of the paving robot is determined based on the motion trajectory and the installation coordinates of each photovoltaic module.

8. The parking method for the paving robot as described in claim 7, characterized in that, Determining the parking position of the paving robot based on the motion trajectory and the installation coordinates of each photovoltaic module includes: Calculate the maximum area that the motion trajectory can cover for each of the installation coordinates, and obtain the target coordinates of the motion trajectory when the coverage area of ​​the motion trajectory on the installation coordinates reaches the maximum. Calculate the relative position of the paving robot and the photovoltaic support when the motion trajectory reaches the target coordinates; The parking position of the paving robot is determined based on the relative position and environmental factors.

9. 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 parking method of the paving robot according to any one of claims 1-8.

10. A paving robot, characterized in that, The paving robot is used to install photovoltaic modules. The paving robot includes a processor, a memory, and a robotic arm. The memory is used to store computer-readable instructions, and the processor is used to call the instructions stored in the memory to execute the method of any one of claims 1-8.

Images