Airborne radar cover laser paint removal path planning method and system

Abstract

The present invention provides a path planning method and system for laser paint removal of an airborne radome, and relates to the field of laser processing and artificial intelligence control technology. The path planning method for laser paint removal of an airborne radome includes: obtaining a binary feature image of a target workpiece to be painted; performing connected domain extraction on the binary feature image to obtain multiple connected domains, and merging all connected domains to obtain multiple merged partitions; performing path planning on each merged partition to obtain corresponding path planning data; performing global planning on the merged partition according to the path planning data of each merged partition to obtain a global composite path planning result. The present invention reduces the complexity of path planning by merging connected domains, and realizes effective scanning path planning only for the expected paint removal area within the range of a regular graphic. On this basis, global planning is used to calculate the optimal access order of multiple merged partitions globally, ensuring that the laser can cover each merged partition and improving the removal efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the field of laser processing and artificial intelligence control technology, and in particular to a method and system for planning a path for laser paint removal of an airborne radome. Background Art

[0002] Aircraft radome is an important part of the aircraft. It is made of fiber-reinforced composite materials. It protects the radar system and ensures the aerodynamic shape of the whole aircraft without affecting the radar performance. In complex working environments, scratches and paint peeling are common types of damage. Scratches are strip-shaped damage caused by collisions with sharp objects, such as sand or accidental collisions with work ladders during flight; paint peeling is mainly caused by aging of the paint layer and decreased adhesion. For scratches and paint peeling damage within the paint layer, the maintenance method is to remove the paint and touch up the paint and other repair processes.

[0003] At present, the paint removal of radomes mostly adopts traditional cleaning methods, such as mechanical, physical, chemical or electrochemical cleaning. Although the operation is simple and the cost is low, there are problems such as damage to the surface matrix, environmental pollution, and reliance on manpower. In addition, in targeted cleaning, due to the problem of operation accuracy, excessive maintenance or even full surface paint removal may occur, resulting in low maintenance efficiency and waste of resources. In order to shorten the maintenance cycle and reduce maintenance costs, a more intelligent and efficient paint removal method needs to be applied. Summary of the invention

[0004] The problem solved by the present invention is one or more of the above-mentioned problems in the prior art.

[0005] In order to solve the above problems, the present invention provides a method and system for airborne radome laser paint removal path planning.

[0006] In a first aspect, the present invention provides a method for planning a path for laser paint removal of an airborne radome, comprising:

[0007] Acquire a binary feature image of the target workpiece to be depainted;

[0008] Extracting connected domains from the binary feature image to obtain multiple connected domains, and merging all the connected domains to obtain multiple merged partitions;

[0009] Performing path planning on each of the merged partitions to obtain corresponding path planning data;

[0010] The merged partitions are globally planned according to the path planning data of each merged partition to obtain a global composite path planning result.

[0011] Optionally, each of the connected domains includes a corresponding coordinate set, and merging all the connected domains to obtain a plurality of merged partitions includes:

[0012] Based on each of the coordinate sets, a coordinate extreme value corresponding to the connected domain is obtained, and according to each of the coordinate extreme values, a regular rectangular boundary corresponding to the connected domain is obtained;

[0013] All the regular rectangular boundaries are merged to obtain a plurality of merged partitions.

[0014] Optionally, merging all the regular rectangular boundaries to obtain a plurality of merged partitions includes:

[0015] Sorting all the regular rectangular boundaries to obtain a set of regular rectangular boundaries;

[0016] The data in the regular rectangular boundary set are merged according to a preset merging criterion to obtain a plurality of merged partitions.

[0017] Optionally, the data in the regular rectangular boundary set are merged according to a preset merging criterion to obtain a plurality of merged partitions, including

[0018] A merging process is performed, the merging process comprising:

[0019] Selecting a first regular rectangular boundary and a target regular rectangular boundary in the current set of regular rectangular boundaries, wherein the target regular rectangular boundary is any other regular rectangular boundary in the current set of regular rectangular boundaries other than the first regular rectangular boundary;

[0020] Determine whether the first regular rectangular boundary and the target regular rectangular boundary meet the preset merging criterion;

[0021] If satisfied, merge the first regular rectangular boundary and the target regular rectangular boundary, and use the merged area data as the first regular rectangular boundary, and return to execute the merging process until all the regular rectangular boundaries in the regular rectangular boundary set are traversed to obtain the merged current rectangular boundary set;

[0022] If not, the next target regular rectangular boundary is selected in sequence, and the step of determining whether the first regular rectangular boundary and the target regular rectangular boundary satisfy the preset merging criterion is returned to, until all the target regular rectangular boundaries in the regular rectangular boundary set are traversed to obtain the current rectangular boundary set;

[0023] The first regular rectangular boundary in the current rectangular boundary set is used as the merged partition, and the remaining regular rectangular boundaries in the current rectangular boundary set constitute the current regular rectangular boundary set, and the merging process is returned to execute, and the above process is repeated until all the merged partitions are obtained.

[0024] Optionally, merging the data in the regular rectangular boundary set according to a preset merging criterion to obtain a plurality of merged partitions includes:

[0025] The merging criterion is:

[0026]

[0027] in, is the midpoint of the X-axis extreme value of the first connected domain, is the midpoint of the Y-axis extreme value of the first connected domain, The midpoint of the x-axis extreme value of the k-th connected domain The midpoint of the Y-axis extreme value of the k-th connected domain are respectively the XY coordinate extreme values ​​of the first connected domain, are respectively the XY coordinate extreme values ​​of the kth connected domain;

[0028] When performing the merging process, when the preset merging criterion is met for the k-1th time, the merged current shot rectangle boundary set is:

[0029]

[0030] Among them, B c is the current rectangular boundary set, k is the serial number sequence of the regular rectangular boundaries, and T is the total number of the regular rectangular boundaries.

[0031] Optionally, performing path planning on each of the merged partitions to obtain corresponding path planning data includes:

[0032] Performing local customized inspired path coverage on each of the merged partitions to obtain the corresponding path planning data;

[0033] The process of selecting the starting point of each path planning data includes:

[0034] Determine the area of ​​the effective target region according to all the merged partitions, and obtain the corresponding static distance;

[0035] Determine the centroid position according to the area of ​​the effective target area and the corresponding static distance, and obtain the corresponding boundary member set based on each of the merged partitions;

[0036] The corresponding starting point is determined according to each of the boundary member sets and the centroid position.

[0037] Optionally, determining the corresponding starting point according to each of the boundary member sets and the centroid position includes:

[0038] Calculating the distance between each member in each boundary member set and the centroid position to obtain a plurality of distance data;

[0039] The member corresponding to the minimum value is selected from all the distance data as the corresponding starting point.

[0040] Optionally, performing path planning on the merged partition according to the path planning data of each merged partition to obtain a global composite path planning result includes:

[0041] Taking each of the merged partitions as a global node, performing access planning on all global nodes based on a composite ant colony algorithm to obtain the global composite path planning result;

[0042] Wherein, the composite ant colony algorithm includes an ant colony cyclic variant model.

[0043] Optionally, performing global planning on the merged partitions according to the path planning data of each merged partition to obtain a global composite path planning result includes:

[0044] By performing path planning on each of the merged partitions, the expected paint removal area in the target workpiece to be painted is effectively scanned to obtain the corresponding path planning data; and through global planning, the optimal access order of each merged partition is calculated according to each path planning data to ensure that the laser accurately covers each merged partition with the shortest distance, thereby obtaining the global composite path planning result.

[0045] In a second aspect, the present invention provides an airborne radome laser paint removal path planning system, comprising:

[0046] An acquisition unit, used for acquiring a binary feature image of a target workpiece to be depainted;

[0047] A merging unit, configured to extract a connected domain from the binary feature image to obtain a plurality of connected domains, and merge all the connected domains to obtain a plurality of merged partitions;

[0048] A planning unit, used for performing global planning on each of the merged partitions to obtain corresponding path planning data;

[0049] The planning unit is further used to perform path planning on the merged partition according to the path planning data of each merged partition to obtain a global composite path planning result.

[0050] The beneficial effects of the path planning method and system for laser paint removal of airborne radar covers of the present invention are as follows: first, an image of the workpiece to be depainted is obtained by image processing technology, and converted into a binary feature image. The purpose of this step is to extract the features of the workpiece surface for subsequent processing. Then the binary feature image is subjected to connected domain extraction to identify different connected areas in the image. These connected domains represent different paint layer areas on the surface of the workpiece. Through connected domain analysis, the interconnected pixels in the image can be grouped into the same group to form multiple connected domains. The extracted multiple connected domains are merged, such as by merging operations according to set merging rules to obtain multiple merged partitions. This step can reduce the complexity of path planning, because the merged area is usually larger than a single connected domain, which is convenient for laser paint removal operations. In the final path planning process, path planning is first performed on each merged partition to generate corresponding path planning data. This process involves path coverage methods with local custom heuristics, such as arch coverage, inner spiral coverage, inward step coverage, etc. The purpose is to obtain the expected coverage path that only scans the paint layer within the merged partition range, so that the subsequent laser scanning and cleaning can obtain a regular rectangular paint removal area corresponding to the number of merges, to ensure that the laser can efficiently cover each merged partition during the paint removal process to avoid omissions or duplications. Finally, based on the path planning data of each merged partition, a global composite path planning result is obtained. The purpose of this step is to integrate the paths of all merged partitions into a continuous laser paint removal path to improve work efficiency and reduce operation time.

[0051] Therefore, the present invention reduces the complexity of path planning by merging connected domains, making the laser paint removal process more efficient and able to complete the task in a shorter time; at the same time, effective scanning path planning only for the expected paint removal area is realized within the range of several merged regular graphics (merged partitions), so that the area to be repainted is regularized, and on this basis, global planning is used to calculate the optimal access order of the global multiple merged partitions to ensure that the laser can cover each merged partition, improve the removal efficiency and accuracy, and provide calculation and operation convenience for subsequent paint repainting and maintenance processes; and optimized path planning can also reduce the use time of the laser, thereby reducing energy consumption and equipment wear, and reducing overall operating costs; and this method can adapt to workpieces of different shapes and sizes, and has strong versatility and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic flow chart of a method for planning a path for laser paint removal of an airborne radome according to an embodiment of the present invention;

[0053] Figure 2 is a schematic diagram of a binary feature image according to an embodiment of the present invention;

[0054] Figure 3A schematic diagram of binary display of all connected domains in an embodiment of the present invention;

[0055] Figure 4 A schematic diagram of the boundaries of the merged partitions according to an embodiment of the present invention;

[0056] Figure 5 A schematic diagram of effective target partitions according to an embodiment of the present invention;

[0057] Figure 6 A schematic diagram of the structure of a path planning system for laser paint removal of an airborne radome according to an embodiment of the present invention;

[0058] Figure 7 It is a schematic diagram of the global composite path planning result of an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0060] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0061] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0062] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0063] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0064] like Figure 1 As shown, an embodiment of the present invention provides a method for planning a path for laser paint removal of an airborne radome, comprising:

[0065] S100, obtaining a binary feature image of a target workpiece to be depainted;

[0066] Specifically, obtaining the binary feature image of the target workpiece to be depainted is the first step in the entire laser depainting path planning process. Use a high-resolution camera or laser scanning equipment to photograph the workpiece to be depainted. Make sure to shoot under good lighting conditions to obtain a clear image. The quality of the image directly affects the effect of subsequent processing. Among them, the target workpiece to be depainted is an airborne radar cover sample with several random irregular shapes of paint damage. The damage of the sample includes scratches within the paint layer and failure phenomena such as paint peeling, which require local paint removal maintenance.

[0067] Preprocess the collected images, including denoising, contrast enhancement, etc. Common denoising methods include median filtering, Gaussian filtering, etc. These methods can effectively remove random noise in the image and improve the accuracy of feature extraction. Convert the color image to a grayscale image. Grayscale processing can simplify the subsequent binarization process, and the threshold segmentation method can be used to convert the grayscale image into a binary image. Common threshold segmentation methods include global threshold method (such as Otsu method) and local threshold method. By setting a threshold, the pixels with grayscale values ​​higher than the threshold are set to white (1), and the pixels with grayscale values ​​lower than the threshold are set to black (0), thereby forming a binary feature image, such as Figure 2 As shown in the figure, the schematic diagram of the binary feature image. In the binary feature image obtained, white represents the paint layer and black represents the damaged area. At this point, the connected domain analysis can be performed to extract all the connected areas to prepare for the subsequent path planning.

[0068] In some embodiments, first, the target workpiece to be partially depainted (radome workpiece) needs to be adjusted and fixed. Assume that the radome has completed the surface block calculation process (the process is to divide the surface approximately into blocks of the plane scanning field, and the size of the block unit depends on the focal depth of the laser beam and the effective scanning range). Next, the graphics acquisition module can be used to sample the block unit area in the field of view, and its feature image is binarized, and vertex calibration is performed at the same time to obtain the spatial coordinate information matching of the binary image matrix.

[0069] It should be noted that the spatial equal arc length projection transformation process is involved in both the sampling and beam scanning processes. However, since the mapping process between the curved surface and the plane is reversible, the two transformation coefficients are approximately equivalent to reverse cancellation. Therefore, in the graphics acquisition and path planning stage and the laser scanning stage, the spatial equal arc length projection transformation process can be omitted once each. In addition, in order to fully demonstrate the path planning process, it is assumed that the scanning unit size is equal to the pixel unit size. At this time, the discrete coordinate system of the path point can be regarded as the pixel coordinate system of the image, that is, the binary feature image includes the coordinate data of multiple path points.

[0070] S200, extracting connected domains from the binary feature image to obtain multiple connected domains, and merging all the connected domains to obtain multiple merged partitions;

[0071] Specifically, in the binary feature image, the white area represents the part to be painted, and the black area represents the part that does not need to be processed. Through the connected domain extraction algorithm (such as depth-first search or breadth-first search), the bwlabel (binary image connected region labeling) function of matlab (Matrix Laboratory) can also be used to label the connected regions in the binary image and return a matrix of the same size as the input binary image, in which each connected region is assigned a unique integer label; all interconnected white pixels in the image can be identified to form multiple connected domains. Each connected domain represents an independent paint layer damage area, such as Figure 3 As shown in the figure, a schematic diagram of the binary display of all connected domains. After extracting multiple connected domains, these connected domains need to be merged. The basis for merging can be the spatial proximity or overlap between connected domains, for example, by setting a merging criterion. By setting a certain merging threshold, adjacent or close connected domains are merged into a larger area to form multiple merged partitions.

[0072] The above process reduces the number of areas that need to be processed by merging connected domains, making the path planning process simpler and more efficient. The merged area is usually larger than a single connected domain, which is convenient for laser paint removal. At the same time, the merged area can reduce the number of laser scans, thereby improving the overall processing efficiency and reducing the operation time. And merging connected domains can ensure that the laser covers all areas that need to be painted during the processing process, avoiding omissions or repeated processing, and improving the quality of paint removal.

[0073] S300, performing path planning on each of the merged partitions to obtain corresponding path planning data;

[0074] Specifically, in the path planning stage, the goal is to generate a laser scanning path for each merged partition to ensure that the laser can effectively cover the entire area to be depainted. Path planning needs to consider the effective scanning range, focal depth, speed, and shape and characteristics of the workpiece. Among them, all merged partitions are covered one by one by a path of a local customized heuristic method, and the coverage method may include arch coverage, inner spiral coverage, inward step coverage, etc.

[0075] Through reasonable path planning, it is ensured that the laser can cover all the paint removal areas in each merged partition to avoid omissions. And effective path planning can reduce the time of laser scanning, improve overall work efficiency, and reduce operating costs. Therefore, by optimizing the path, reducing unnecessary movement and repeated scanning, reducing the wear and energy consumption of laser equipment, and reasonable path planning can ensure uniform laser irradiation and improve the quality and consistency of paint removal.

[0076] S400, performing global planning on the merged partition according to the path planning data of each merged partition to obtain a global composite path planning result.

[0077] Specifically, the path planning data of each merged partition can be used to extract the starting and ending path nodes of the path covered by each sub-partition. Using the idea of ​​composite path planning, each merged partition is regarded as a number of global nodes for subsequent global access order planning. A global path planning algorithm (such as simulated annealing, greedy algorithm, ant colony algorithm, particle swarm algorithm, genetic algorithm, etc.) is used to integrate the path data of each merged partition. The goal of global path planning is to find the optimal path from the starting point to all merged partitions to ensure that the laser equipment can efficiently cover all areas to be processed. During the integration process, it is necessary to consider the distance between the merged partitions, the connectivity of the path, and the motion limitations of the laser equipment. After generating the global composite path, the path can be further optimized to reduce unnecessary movement and repeated scanning. For example, the overall path can be optimized by merging adjacent path segments, adjusting the path order, etc., and the global composite path planning result data finally generated will include the coordinates of all path points.

[0078] By integrating the path planning data of each merged partition, the moving distance and time of the laser equipment can be reduced, and the overall operation efficiency can be improved; at the same time, the global composite path planning results ensure that all areas to be processed can be fully irradiated by the laser to avoid omissions; and the optimized path reduces unnecessary movement, reducing the energy consumption and wear of the laser equipment. Finally, reasonable path planning can ensure uniform laser irradiation and improve the quality and consistency of paint removal.

[0079] In this embodiment, first, an image of the workpiece to be depainted is obtained by image processing technology, and converted into a binary feature image. The purpose of this step is to extract the features of the workpiece surface for subsequent processing. Then, the connected domain is extracted from the binary feature image to identify different connected areas in the image. These connected domains represent different paint layer areas on the surface of the workpiece. Through connected domain analysis, the interconnected pixels in the image can be grouped into the same group to form multiple connected domains. The extracted multiple connected domains are merged, such as by merging according to the set merging rules to obtain multiple merged partitions. This step can reduce the complexity of path planning, because the merged area is usually larger than a single connected domain, which is convenient for laser paint removal operations. In the final path planning process, path planning is first performed on each merged partition to generate corresponding path planning data. This process involves path coverage methods with local custom heuristics, such as arch coverage, inner spiral coverage, inward step coverage, etc. The purpose is to obtain the expected coverage path that only scans the paint layer within the merged partition range, so that the subsequent laser scanning and cleaning can obtain a regular rectangular paint removal area corresponding to the number of merges, to ensure that the laser can efficiently cover each merged partition during the paint removal process to avoid omissions or duplications. Finally, based on the path planning data of each merged partition, a global composite path planning result is obtained. The purpose of this step is to integrate the paths of all merged partitions into a continuous laser paint removal path to improve work efficiency and reduce operation time.

[0080] Therefore, this embodiment reduces the complexity of path planning by merging connected domains, making the laser paint removal process more efficient and able to complete the task in a shorter time; at the same time, effective scanning path planning is implemented only for the expected paint removal area within the range of several merged regular graphics (merged partitions), so that the area to be repainted is regularized, and on this basis, global planning is used to calculate the optimal access order of the global multiple merged partitions to ensure that the laser can accurately cover each merged partition with the shortest distance, improve the removal efficiency and accuracy, and provide calculation and operation convenience for subsequent paint repainting and maintenance processes; and optimized path planning can also reduce the use time of the laser, thereby reducing energy consumption and equipment wear, and reducing overall operating costs; and this method can adapt to workpieces of different shapes and sizes, and has strong versatility and adaptability.

[0081] Optionally, each of the connected domains includes a corresponding coordinate set, and merging all the connected domains to obtain a plurality of merged partitions includes:

[0082] Based on each of the coordinate sets, a coordinate extreme value corresponding to the connected domain is obtained, and according to each of the coordinate extreme values, a regular rectangular boundary corresponding to the connected domain is obtained;

[0083] All the regular rectangular boundaries are merged to obtain a plurality of merged partitions.

[0084] Specifically, the connected domain is extracted from the binary feature image, and the bwlabel function of MATLAB can be used to better implement the connected domain extraction process. The bwlabel function realizes the segmentation of each separate connected interval by setting the connectivity judgment mode (4-connected or 8-connected). This function is applied in the invention to obtain the independent distribution of each scratch or paint peeling area, that is, to extract the coordinate set of each connected domain.

[0085] For each connected domain's coordinate set, calculate its coordinate extrema, including the minimum and maximum x and y coordinates. Based on the calculated coordinate extrema, generate a regular rectangular boundary for each connected domain. The four corner coordinates of the rectangle can be expressed as: upper left corner: (min_x, min_y), lower right corner: (max_x, max_y); these rectangular boundaries provide the basis for the subsequent merging steps. All regular rectangular boundaries constitute a regular rectangular boundary set, which can be expressed as:

[0086]

[0087] Among them, k is the numbering sequence of the initial connected domain. In the multi-objective iterative merging process, k is recorded as the real-time merged partition numbering sequence. T is the total number of connected domains extracted in the initial state. In the multi-objective iterative merging process, T is recorded as the real-time total number of merged partitions. are the XY coordinate extreme values ​​of the first connected domain, are the XY coordinate extreme values ​​of the kth connected domain, B (k) A set of regular rectangular boundaries.

[0088] By merging multiple connected domains, the subsequent path planning and processing steps can be simplified and the computational complexity can be reduced. The merged partitions reduce the number of laser device movements and improve the overall operation efficiency. At the same time, the merged partitions ensure that all areas to be processed can be fully irradiated by the laser to avoid omissions. The regular rectangular boundaries are easy to manage and process, which is convenient for subsequent path planning and laser control.

[0089] Therefore, through the above process, the coordinate set of the connected domain is used to generate a regular rectangular boundary and merge them, which not only improves the processing efficiency, but also ensures the coverage rate during the laser paint removal process, providing effective support for subsequent path planning.

[0090] Optionally, merging all the regular rectangular boundaries to obtain a plurality of merged partitions includes:

[0091] Sorting all the regular rectangular boundaries to obtain a set of regular rectangular boundaries;

[0092] The data in the regular rectangular boundary set are merged according to a preset merging criterion to obtain a plurality of merged partitions.

[0093] Optionally, the data in the regular rectangular boundary set are merged according to a preset merging criterion to obtain a plurality of merged partitions, including

[0094] A merging process is performed, the merging process comprising:

[0095] Selecting a first regular rectangular boundary and a target regular rectangular boundary in the current set of regular rectangular boundaries, wherein the target regular rectangular boundary is any other regular rectangular boundary in the current set of regular rectangular boundaries other than the first regular rectangular boundary;

[0096] Determine whether the first regular rectangular boundary and the target regular rectangular boundary meet the preset merging criterion;

[0097] If satisfied, merge the first regular rectangular boundary and the target regular rectangular boundary, and use the merged area data as the first regular rectangular boundary, and return to execute the merging process until all the regular rectangular boundaries in the regular rectangular boundary set are traversed to obtain the merged current rectangular boundary set;

[0098] If not, the next target regular rectangular boundary is selected in sequence, and the step of determining whether the first regular rectangular boundary and the target regular rectangular boundary satisfy the preset merging criterion is returned to, until all the target regular rectangular boundaries in the regular rectangular boundary set are traversed to obtain the current rectangular boundary set;

[0099] The first regular rectangular boundary in the current rectangular boundary set is used as the merged partition, and the remaining regular rectangular boundaries in the current rectangular boundary set constitute the current regular rectangular boundary set, and the merging process is returned to execute, and the above process is repeated until all the merged partitions are obtained.

[0100] Optionally, merging the data in the regular rectangular boundary set according to a preset merging criterion to obtain a plurality of merged partitions further comprises:

[0101] The merging criterion is:

[0102]

[0103] in, is the midpoint of the X-axis extreme value of the first connected domain, is the midpoint of the Y-axis extreme value of the first connected domain, The midpoint of the x-axis extreme value of the k-th connected domain The midpoint of the Y-axis extreme value of the k-th connected domain are respectively the XY coordinate extreme values ​​of the first connected domain, are respectively the XY coordinate extreme values ​​of the kth connected domain;

[0104] When performing the merging process, when the preset merging criterion is met for the k-1th time, the merged current shot rectangle boundary set is:

[0105]

[0106] Among them, B c is the current rectangular boundary set, k is the serial number sequence of the regular rectangular boundaries, and T is the total number of the regular rectangular boundaries.

[0107] After the merging based on the first regular rectangular boundary is completed, the final current rectangular boundary set is obtained, the first regular rectangular boundary in the current rectangular boundary set is used as a merged partition, the first regular rectangular boundary is removed from the current rectangular boundary set, and the remaining regular rectangular boundaries are updated in sequence, that is, the serial numbers are all moved forward one, at this time, the current first regular rectangular boundary is again judged with the boundaries of other regular rectangular boundaries in the current boundary set by the preset merging criterion, that is, the merging process is repeated based on the current first regular rectangular boundary, after the merging process is completed, it is used as another merged partition, and is removed from the current current rectangular boundary set at this time, and the above process is repeated until all merge partition operations are completed, and all merged partitions are obtained, which can be expressed as: Wherein, m=1...N, and N is the number of the final merged partitions.

[0108] In some embodiments, during the beat-by-beat merging process, the following iterative assignment processing is performed for different merging situations:

[0109]

[0110]

[0111] Update equations (1)-(7) iteratively one by one to obtain multiple merged partitions, which include a set of multi-target partition boundaries (such as Figure 4 , the boundary diagram of the merged partition) and the result of the rule-merged partition (such as Figure 5 , Schematic diagram of effective target partition), Figure 4 The white part in the middle represents the paint layer, the black part represents the damaged area, and the dashed box represents the calculated merged boundary. Figure 5 The final regular rectangular merging area is displayed, and the intact paint layer part that does not need to be removed is placed as the non-target background area.

[0112] In this embodiment, the white area represents the paint layer. It should be noted that the method of the present invention mainly plans and scans the paint layer portion within the virtual frame. Therefore, the white area within the virtual frame is regarded as the target area defined in the present invention.

[0113] By laser cleaning the target area, the subsequent process will form a complete and regular rectangular paint removal area. This planning method not only improves the cleaning efficiency, but also facilitates the subsequent calculation and processing, ensuring the smooth progress of the subsequent process.

[0114] Through the above process, the coordinate set of the connected domain is used to generate regular rectangular boundaries and merge them, which not only improves the processing efficiency but also ensures the coverage rate during laser paint removal, providing effective support for subsequent path planning.

[0115] Optionally, performing path planning on each of the merged partitions to obtain corresponding path planning data includes:

[0116] Performing local customized inspired path coverage on each of the merged partitions to obtain the corresponding path planning data;

[0117] The process of selecting the starting point of each path planning data includes:

[0118] Determine the area of ​​the effective target region according to all the merged partitions, and obtain the corresponding static distance;

[0119] Determine the centroid position according to the area of ​​the effective target area and the corresponding static distance, and obtain the corresponding boundary member set based on each of the merged partitions;

[0120] The corresponding starting point is determined according to each of the boundary member sets and the centroid position.

[0121] Specifically,

[0122] Optionally, determining the corresponding starting point according to each of the boundary member sets and the centroid position includes:

[0123] Calculating the distance between each member in each boundary member set and the centroid position to obtain a plurality of distance data;

[0124] The member corresponding to the minimum value is selected from all the distance data as the corresponding starting point.

[0125] Specifically, when performing local customized heuristic path coverage on all merged partitions one by one, the method of selecting the starting point is not fixed. In order to consider the impact of the starting point position on the global partition access process, the present invention proposes a centroid-based starting point selection method. Specifically, first calculate the centroid position of the area to be painted removed in the binary feature image as a reference point. Then, in each merged partition, traverse its paint layer boundary elements to find the boundary point closest to the centroid. This point will be marked as the path starting point of the current partition. Through this method, the starting point can be effectively selected, thereby optimizing the path coverage process, improving overall work efficiency, and ensuring comprehensive coverage of each merged partition.

[0126] First, calculate the centroid of the paint layer area to be removed (all the merged partitions). The area of ​​the effective target area for laser paint removal needs to be calculated as A, the differential is dA, and for the xy coordinates of the effective area, the static distances of A to the x-axis and y-axis are S respectively. x =∫ydA,S y =∫xdA, then the centroid position coordinates are expressed as: X c and Y c are the horizontal and vertical coordinates of the centroid respectively.

[0127] Select the merge partition boundary member set B r W = ∪ border B r W border (Among them, B r W is the boundary set of the rth merged partition, r is a counting variable, B r W bprder is the independent unit at the border of the merged sub-partition, border is a counting variable, U border Refers to the collection), traverses and calculates the distance between the coordinates of each member point in the merged partition and the centroid position, and selects the member point closest to the centroid as the starting point of the heuristic path of the merged partition. Then, the end path point is determined by calculating the covering path.

[0128] On this basis, the concept of composite path planning is adopted, and each merged partition is regarded as a number of global nodes for subsequent global access sequence planning. It should be noted that the access sequence and access path between partitions are not equivalent to the laser scanning path, but serve as a guide for the series connection of global partitions. During the migration access process between partitions, the state of the laser spot must be controlled to be disabled to prevent the laser spot from having unnecessary effects on non-paint removal target areas. In this way, path planning can be effectively optimized to ensure the accuracy and safety of laser paint removal operations.

[0129] Optionally, performing path planning on the merged partition according to the path planning data of each merged partition to obtain a global composite path planning result includes:

[0130] Taking each of the merged partitions as a global node, performing access planning on all global nodes based on a composite ant colony algorithm to obtain the global composite path planning result;

[0131] Wherein, the composite ant colony algorithm includes an ant colony cyclic variant model.

[0132] Specifically, the path planning for the merged partition is performed according to the path planning data of each merged partition, that is, the planning method of the global access order may include simulated annealing, greedy algorithm, ant colony algorithm, particle swarm algorithm, genetic algorithm, etc.

[0133] After completing the heuristic path planning within the merged partition, combined with the idea of ​​composite planning, all merged partitions are considered as a whole, which is used as the access node for external global access. In this whole, the representative node includes the starting point and the end point of the heuristic path within the merged partition.

[0134] During the access process, the access point can be located at the starting point or the end point of the coverage path within the partition. By connecting the coverage paths within the merged partitions in series, the next optimal access node can be accessed effectively. This method not only optimizes path planning, but also ensures effective connection and smooth transition between partitions during the global access process.

[0135] In some embodiments, in the process of planning the global access sequence, the intelligent method used is a composite ant colony algorithm that integrates local path information. The algorithm uses a variant model of Ant-Cycle (ant colony cycle variant model) to update the overall pheromone increment. Specifically, the pheromone increment can be expressed as:

[0136]

[0137] in, It is expressed as the pheromone increment of the μth ant left between the nodes of the access path (i, j) at time t+n, Q is the ant pheromone intensity factor, L μ is the length of the access path that the μth ant crosses the node in this cycle.

[0138] By setting appropriate parameters corresponding to the ant colony algorithm and combining the processing steps of this embodiment, a composite path planning effect (global composite path planning result) is finally obtained. Figure 6As shown, a schematic diagram of the global composite path planning result (schematic diagram of the composite path planning effect) is shown. In the figure, the paint surface areas outside the merged partitions that do not need to be depainted are represented in black, the natural scratches or paint peeling areas within each merged partition are represented in white, and the expected local scanning coverage path within each merged partition is represented by gray lines in the rectangular box, and the global access process between the merged partitions is represented by dotted lines.

[0139] Through the analysis of this example, the laser scanning planning of the radome sample for random multi-targets in the field of view is obtained. At the same time, the calculation results show that the access distance between global partitions is 23, the total path length of local coverage is 29165, and the total distance of the composite path is 29188. These results effectively reflect the application effect of the proposed method in path planning.

[0140] Optionally, performing global planning on the merged partitions according to the path planning data of each merged partition to obtain a global composite path planning result includes:

[0141] By performing path planning on each of the merged partitions, the expected paint removal area in the target workpiece to be painted is effectively scanned to obtain the corresponding path planning data; and through global planning, the optimal access order of each merged partition is calculated according to each path planning data to ensure that the laser accurately covers each merged partition with the shortest distance, thereby obtaining the global composite path planning result.

[0142] like Figure 7 As shown, an embodiment of the present invention provides an airborne radome laser paint removal path planning system, comprising:

[0143] An acquisition unit, used for acquiring a binary feature image of a target workpiece to be depainted;

[0144] A merging unit, configured to extract a connected domain from the binary feature image to obtain a plurality of connected domains, and merge all the connected domains to obtain a plurality of merged partitions;

[0145] A planning unit, used to perform path planning on each of the merged partitions to obtain corresponding path planning data;

[0146] The planning unit is further used to perform global planning on the merged partitions according to the path planning data of each merged partition to obtain a global composite path planning result.

[0147] The airborne radome laser paint removal path planning system of this embodiment is used to implement the airborne radome laser paint removal path planning method as described above. Its advantages over the prior art are the same as the advantages of the above-mentioned airborne radome laser paint removal path planning method over the prior art, and will not be repeated here.

[0148] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A method for planning a path for laser paint removal of an airborne radome, characterized in that: include: Acquire a binary feature image of the target workpiece to be depainted; Performing connected domain extraction on the binary feature image to obtain multiple connected domains, each of the connected domains includes a corresponding coordinate set; All the connected domains are merged to obtain multiple merged partitions, including: Based on each of the coordinate sets, a coordinate extreme value corresponding to the connected domain is obtained, and according to each of the coordinate extreme values, a regular rectangular boundary corresponding to the connected domain is obtained; Merging all the regular rectangular boundaries to obtain a plurality of merged partitions includes: Sorting all the regular rectangular boundaries to obtain a set of regular rectangular boundaries; Merging the data in the regular rectangular boundary set according to a preset merging criterion to obtain a plurality of merged partitions, including: A merging process is performed, the merging process comprising: Selecting a first regular rectangular boundary and a target regular rectangular boundary in the current set of regular rectangular boundaries, wherein the target regular rectangular boundary is any other regular rectangular boundary in the current set of regular rectangular boundaries other than the first regular rectangular boundary; Determine whether the first regular rectangular boundary and the target regular rectangular boundary meet the preset merging criterion; If satisfied, merge the first regular rectangular boundary and the target regular rectangular boundary, and use the merged area data as the first regular rectangular boundary, and return to execute the merging process until all the regular rectangular boundaries in the regular rectangular boundary set are traversed to obtain the merged current rectangular boundary set; If not, the next target regular rectangular boundary is selected in sequence, and the step of determining whether the first regular rectangular boundary and the target regular rectangular boundary satisfy the preset merging criterion is returned to, until all the target regular rectangular boundaries in the regular rectangular boundary set are traversed to obtain the current rectangular boundary set; The first regular rectangular boundary in the current rectangular boundary set is used as the merged partition, and the remaining regular rectangular boundaries in the current rectangular boundary set constitute the current regular rectangular boundary set, and the merging process is returned to be executed, and the above process is repeated until all the merged partitions are obtained; Wherein, the preset merging criteria include: The preset merging criteria are: in, is the midpoint of the X-axis extreme value of the first connected domain, is the midpoint of the Y-axis extreme value of the first connected domain, The midpoint of the x-axis extreme value of the k-th connected domain The midpoint of the Y-axis extreme value of the k-th connected domain are respectively the XY coordinate extreme values ​​of the first connected domain, are respectively the XY coordinate extreme values ​​of the kth connected domain; When performing the merging process, when the preset merging criterion is met for the k-1th time, the merged current shot rectangle boundary set is: Among them, B c is the current rectangular boundary set, k is the serial number sequence of the regular rectangular boundaries, and T is the total number of the regular rectangular boundaries; Performing path planning on each of the merged partitions to obtain corresponding path planning data; The merged partitions are globally planned according to the path planning data of each merged partition to obtain a global composite path planning result.

2. The method for airborne radome laser paint removal path planning according to claim 1, characterized in that: The performing path planning on each of the merged partitions to obtain corresponding path planning data includes: Performing local customized inspired path coverage on each of the merged partitions to obtain the corresponding path planning data; The process of selecting the starting point of each path planning data includes: Determine the area of ​​the effective target area according to all the merged partitions, and obtain the corresponding static distance; Determine the centroid position according to the area of ​​the effective target area and the corresponding static distance, and obtain the corresponding boundary member set based on each of the merged partitions; The corresponding starting point is determined according to each of the boundary member sets and the centroid position.

3. The method for airborne radar cover laser paint removal path planning according to claim 2, characterized in that: The determining the corresponding starting point according to each of the boundary member sets and the centroid position includes: Calculating the distance between each member in each boundary member set and the centroid position to obtain a plurality of distance data; The member corresponding to the minimum value among all the distance data is selected as the corresponding starting point.

4. The method for airborne radome laser paint removal path planning according to claim 1, characterized in that: The performing path planning on the merged partition according to the path planning data of each merged partition to obtain a global composite path planning result includes: Taking each of the merged partitions as a global node, performing access planning on all global nodes based on a composite ant colony algorithm to obtain the global composite path planning result; Wherein, the composite ant colony algorithm includes an ant colony cyclic variant model.

5. The method for airborne radome laser paint removal path planning according to any one of claims 1 to 4, characterized in that: The globally planning the merged partitions according to the path planning data of each merged partition to obtain a global composite path planning result includes: By performing path planning for each of the merged partitions, the expected paint removal area in the target workpiece to be depainted is effectively scanned to obtain the corresponding path planning data; and through global planning, the optimal access order of each merged partition is calculated according to each path planning data to ensure that the laser accurately covers each merged partition with the shortest distance, thereby obtaining the global composite path planning result.

6. An airborne radome laser paint removal path planning system, characterized in that: include: An acquisition unit, used for acquiring a binary feature image of a target workpiece to be depainted; A merging unit, used for performing connected domain extraction on the binary feature image to obtain multiple connected domains, each of which includes a corresponding coordinate set; Merging all the connected domains to obtain a plurality of merged partitions, including: obtaining a coordinate extreme value corresponding to the connected domain based on each coordinate set, and obtaining a regular rectangular boundary corresponding to the connected domain according to each coordinate extreme value; merging all the regular rectangular boundaries to obtain a plurality of merged partitions, including: sorting all the regular rectangular boundaries to obtain a regular rectangular boundary set; merging each data in the regular rectangular boundary set according to a preset merging criterion to obtain a plurality of merged partitions, including: executing a merging process, the merging process including: Selecting a first regular rectangular boundary and a target regular rectangular boundary in the current set of regular rectangular boundaries, wherein the target regular rectangular boundary is any other regular rectangular boundary in the current set of regular rectangular boundaries other than the first regular rectangular boundary; Determine whether the first regular rectangular boundary and the target regular rectangular boundary meet the preset merging criterion; If satisfied, merge the first regular rectangular boundary and the target regular rectangular boundary, and use the merged area data as the first regular rectangular boundary, and return to execute the merging process until all the regular rectangular boundaries in the regular rectangular boundary set are traversed to obtain the merged current rectangular boundary set; If not, the next target regular rectangular boundary is selected in sequence, and the step of determining whether the first regular rectangular boundary and the target regular rectangular boundary satisfy the preset merging criterion is returned to, until all the target regular rectangular boundaries in the regular rectangular boundary set are traversed to obtain the current rectangular boundary set; The first regular rectangular boundary in the current rectangular boundary set is used as the merged partition, and the remaining regular rectangular boundaries in the current rectangular boundary set constitute the current regular rectangular boundary set, and the merging process is returned to be executed, and the above process is repeated until all the merged partitions are obtained; Wherein, the preset merging criteria include: The preset merging criteria are: in, is the midpoint of the X-axis extreme value of the first connected domain, is the midpoint of the Y-axis extreme value of the first connected domain, The midpoint of the x-axis extreme value of the k-th connected domain The midpoint of the Y-axis extreme value of the k-th connected domain are respectively the XY coordinate extreme values ​​of the first connected domain, are respectively the XY coordinate extreme values ​​of the kth connected domain; When performing the merging process, when the preset merging criterion is met for the k-1th time, the merged current shot rectangle boundary set is: Among them, B c is the current rectangular boundary set, k is the serial number sequence of the regular rectangular boundaries, and T is the total number of the regular rectangular boundaries; A planning unit, used to perform path planning on each of the merged partitions to obtain corresponding path planning data; The planning unit is further used to perform global planning on the merged partitions according to the path planning data of each merged partition to obtain a global composite path planning result.

Images