Laser cleaning path planning method and system and electronic equipment

Abstract

The invention provides a laser cleaning path planning method and system and electronic equipment, and relates to the technical field of laser surface treatment.The method comprises the steps that imaging collection is conducted on the surface of a to-be-cleaned workpiece, coordinate registration and binarization processing are conducted, and a binarization image of the surface of the to-be-cleaned workpiece is obtained; regularly partitioning the surface of the to-be-cleaned workpiece according to the binary image to obtain a plurality of sub-partitions, and performing path coverage on each sub-partition to obtain a local path of each sub-partition; mapping a starting point of the local path of each sub-partition to obtain a global representative point set of all the local paths, and generating a traveling salesman problem model according to the global representative point set; solving the traveling salesman problem model through a native heterogeneous competitive-cooperative ant colony optimization algorithm to obtain an optimal global access sequence of all local paths; and generating a laser cleaning scanning path of the surface of the to-be-cleaned workpiece. According to the invention, the sensitivity of path planning is improved, and finally a high-quality laser cleaning scanning path is generated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of laser surface treatment, in particular to a laser cleaning path planning method and system and electronic equipment. BACKGROUND

[0002] With the transformation of modern industry to high-end, intelligent and green, the requirements for surface cleanliness, structural integrity and operation reliability of key components and core equipment are increasingly stringent. As an advanced surface treatment method, cleaning technology has been applied to multi-level precision selective cleaning with flexible regulation and control characteristics, combined with machine vision and intelligent optimization algorithm. Currently, traditional ant colony algorithm is usually used for laser cleaning path planning.

[0003] In related technologies, due to the excessive dependence of traditional ant colony algorithm on hyperparameters such as pheromone heuristic factor, technical personnel need to manually configure and repeatedly debug based on experience, which is not only subjective and time-consuming, but also has great limitations in hyperparameter adaptation scenarios. When the distribution of contaminants on the workpiece surface changes during laser cleaning and the cleaning accuracy requirement is adjusted, the path planning is difficult to quickly respond to dynamic working conditions, resulting in low cleaning efficiency, which seriously restricts the large-scale application of laser intelligent cleaning technology. SUMMARY

[0004] The problem solved by the present application is how to improve the path planning sensitivity under the ant colony algorithm.

[0005] To solve the above problems, the present application provides a laser cleaning path planning method, system and electronic equipment.

[0006] In a first aspect, the laser cleaning path planning method of the present application comprises: imaging and collecting the surface of the workpiece to be cleaned, and performing coordinate registration and binary processing to obtain a binary image of the surface of the workpiece to be cleaned; According to the binary image, the surface of the workpiece to be cleaned is regularly partitioned to obtain a plurality of sub-partitions, and the path coverage is performed for each sub-partition to obtain a local path of each sub-partition; mapping the starting point of the local path of each sub-partition to obtain a global representative point set of all local paths, and generating a traveling salesman problem model according to the global representative point set; solving the traveling salesman problem model by using a native heterogeneous competition-cooperation ant colony optimization algorithm to obtain an optimal global access sequence of all local paths; According to the optimal global access sequence, a laser cleaning scanning path of the surface of the workpiece to be cleaned is generated.

[0007] Optionally, the step of imaging and acquiring the surface of the workpiece to be cleaned, performing coordinate registration and binarization processing to obtain a binarized image of the surface of the workpiece to be cleaned includes: Digital imaging is performed on the surface of the workpiece to be cleaned to obtain the original grayscale image of the surface of the workpiece to be cleaned; The original grayscale image is registered with the actual spatial position of the surface of the workpiece to be cleaned by using a coordinate transformation algorithm to obtain a registered grayscale image. Adaptive thresholding or maximum inter-class variance segmentation is performed on the registered grayscale image to obtain the binarized image of the surface of the workpiece to be cleaned.

[0008] Optionally, the step of regularly partitioning the surface of the workpiece to be cleaned based on the binarized image to obtain multiple sub-partitions includes: Connectivity analysis is performed on the foreground pixels in the binarized image to obtain multiple connected components to be cleaned; For each of the connected components to be cleaned, extract the minimum bounding contour region of the connected component to be cleaned; The threshold for partition size is determined based on the scanning range of the field lens of the laser cleaning device, the cleaning accuracy requirements, and the path planning efficiency requirements. Based on the partition size threshold, the connected component to be cleaned is adaptively divided to obtain multiple sub-partitions.

[0009] Optionally, the step of performing path overlay for each of the sub-partitions to obtain the local path of each sub-partition includes: The scanning reference direction of the local path is determined based on the shape of the sub-partition and the adaptability of the scanning direction of the laser cleaning device. Based on the scanning reference direction and the path coverage algorithm corresponding to the shape of the sub-partition, the starting point, ending point, and path node sequence in the sub-partition are obtained; An initial local path is generated for each sub-partition based on the starting point, the ending point, and the path node sequence in the sub-partition. The initial local path is optimized by using a Bézier curve to obtain the local path.

[0010] Optionally, mapping the starting point of the local path in each of the sub-partitions to obtain a global representative point set for all the local paths includes: Obtain the local coordinates of the starting point of each sub-partition in the local coordinate system of the sub-partition; Based on the local coordinates of the starting point, determine the coordinate registration relationship between the image pixels of the sub-partition and their actual spatial positions; According to the coordinate registration relationship, the local coordinates of the starting points are converted into global coordinates in a global coordinate system; The global coordinates of the starting points of all sub-partitions are summarized to form the global representative point set.

[0011] Optionally, the generating of the traveling salesman problem model according to the global representative point set comprises: Each of the starting points in the global representative point set is taken as a node in the traveling salesman problem model, and the Euclidean distance between any two nodes is calculated; The Euclidean distance is taken as the path cost between the two nodes, and all the nodes are traversed to determine the total path cost; The traveling salesman problem model containing a node set, a path cost matrix and a traversal constraint condition is constructed with the minimum total path cost as the optimization objective.

[0012] Optionally, the solving of the traveling salesman problem model by the native isomerism competition-cooperation ant colony optimization algorithm obtains an optimal global access sequence of all the local paths, comprising: According to the preset regulation threshold and preset grouping step coefficient of the native isomerism competition stage, an expected heuristic factor parameter mapping table is constructed according to the ant colony population size, initial pheromone concentration, pheromone volatility factor and native isomerism competition stage; In the native isomerism competition, the initial pheromone concentration is kept unchanged, and the expected heuristic factor is sequentially injected into the ant colony population in order according to the parameter mapping table to drive the ants to independently complete path construction, and the total path cost, the first-order difference and the second-order difference of the total path cost are obtained; When the iteration number of the ant colony population reaches the preset regulation threshold, the second-order difference is classified by support vector machine to obtain a fluctuation convergence segment, and the expected heuristic factor corresponding to the minimum first-order difference in the fluctuation convergence segment is taken as the optimal expected heuristic factor; According to the optimal expected heuristic factor, the ant colony iteration is performed in combination with the probability transfer algorithm to obtain the optimal global access sequence.

[0013] Optionally, the generating of the laser cleaning scanning path of the surface of the workpiece to be cleaned according to the optimal global access sequence comprises: The nodes in the optimal global access sequence are sequentially mapped to the execution order of the nodes corresponding to the sub-partitions; The local paths of each of the sub-partitions are sequentially extracted according to the execution order of the sub-partitions, and the local paths are spliced to obtain a spliced overall path; The spliced overall path is subjected to speed planning to generate the laser cleaning scanning path.

[0014] In a second aspect, a laser cleaning path planning system is provided, comprising: An imaging acquisition processing unit is configured to acquire and process images of a surface of a workpiece to be cleaned, to perform coordinate registration and binarization, and to obtain a binarized image of the surface of the workpiece to be cleaned; A local path determination unit is configured to divide the surface of the workpiece to be cleaned according to the binarized image, to obtain a plurality of sub-zones, and to determine a local path for each of the sub-zones; A modeling unit is configured to map starting points of the local paths of each of the sub-zones to obtain a global representative point set of all the local paths, and to generate a traveling salesman problem model according to the global representative point set; A solving unit is configured to solve the traveling salesman problem model by using a native heterogeneous competition-cooperation ant colony optimization algorithm to obtain an optimal global access sequence of all the local paths; A path generation unit is configured to generate a laser cleaning scanning path for the surface of the workpiece to be cleaned according to the optimal global access sequence.

[0015] In a third aspect, an electronic device is provided, comprising a memory and a processor; The memory is configured to store a computer program; The processor is configured to implement the laser cleaning path planning method as described above when executing the computer program.

[0016] The laser cleaning path planning method, system and electronic equipment can accurately obtain the distribution information of the contaminants on the workpiece surface by imaging and collecting the surface of the workpiece to be cleaned and performing coordinate registration and binarization processing. The coordinate registration ensures the accurate correspondence between the image and the actual workpiece surface position, and the binarization processing simplifies the image into a contaminant region and a non-contaminant region, highlighting the key cleaning area and providing a clear basis for subsequent zoning and path planning. According to the binarized image, the workpiece surface is regularly zoned to obtain a plurality of sub-zones, and a local path is generated for each sub-zone. Dividing the workpiece surface into a plurality of sub-zones helps to decompose the complex cleaning task into a plurality of relatively simple sub-tasks. This decomposition method can reduce the complexity of the problem, and generating a local path for each sub-zone can ensure that the contaminants in each region are effectively cleaned. For example, for a workpiece surface with irregular shape and uneven contaminant distribution, regular zoning can divide it into a plurality of small regions that are easy to handle, and the local path in each small region can be designed flexibly according to the contamination situation of the region, improving the pertinence and efficiency of cleaning. Then, the starting points of the local paths are mapped to global representative points, the problem is converted into a traveling salesman problem, and the traveling salesman solving method is used to find the optimal global access sequence. The present application makes full use of the optimization theory and algorithm of the traveling salesman problem to provide an effective framework for solving the laser cleaning path planning problem.

[0017] The traveling salesman problem model is solved by the original heterogeneous competition-cooperation ant colony optimization algorithm to obtain the optimal global access sequence of all local paths. Compared with the traditional ant colony algorithm, the original heterogeneous competition-cooperation ant colony optimization algorithm reduces the over-reliance on hyperparameters by introducing a competition-cooperation mechanism. During the algorithm running process, different ant colonies can compete and cooperate with each other, sharing information, so as to converge to the optimal solution faster. This algorithm can quickly re-plan the path under dynamic working conditions such as changes in the distribution of contaminants on the workpiece surface or adjustments in cleaning precision requirements, improving the sensitivity of laser cleaning path planning. For example, when new contaminated areas appear on the workpiece surface or the contamination level of some areas increases, the algorithm can adjust the global access sequence in time to ensure that the cleaning path can adapt to the new situation, thereby improving the cleaning efficiency. According to the optimal global access sequence, a laser cleaning scanning path for the workpiece surface to be cleaned is generated. The finally generated laser cleaning scanning path considers the local paths of each sub-zone and the global optimal access sequence, and can realize efficient and accurate laser cleaning. This path planning method can ensure that the laser cleaning equipment accesses each contaminated area in the optimal order during the cleaning process, reduces the invalid movement of the equipment, improves the cleaning efficiency, solves the problem of low sensitivity of traditional ant colony algorithm path planning and difficulty in adapting to dynamic working conditions, and effectively promotes the large-scale application of laser intelligent cleaning technology.

[0018] In general, the present application effectively solves the deficiencies of the traditional ant colony algorithm in dynamic path planning, greatly improves the sensitivity of path planning, and enables the entire laser cleaning path planning method to efficiently adapt to different working conditions, ultimately generating a high-quality laser cleaning scanning path. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 A flowchart of the laser cleaning path planning method in the embodiment of the present application is shown in the figure. Figure 2 A target feature binary image in the embodiment of the present application is shown in the figure. Figure 3 A target partition and local path coverage in the embodiment of the present application is shown in the figure. Figure 4 A parameter competition curve under the regulation of ant colony prokaryotic heteromorphism in the embodiment of the present application is shown in the figure. Figure 5 A differential change amount feature curve in the embodiment of the present application is shown in the figure. Figure 6 An hyperparameter extraction result in the embodiment of the present application is shown in the figure. Figure 7 A dynamic self-adaptive optimization curve based on the ant colony prokaryotic heteromorphism competition-cooperation mechanism in the embodiment of the present application is shown in the figure. Figure 8 A global self-adaptive planning effect in the embodiment of the present application is shown in the figure. Figure 9 A structure diagram of the laser cleaning path planning system in the embodiment of the present application is shown in the figure. Figure 10 A structure diagram of the electronic device in the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0020] In order to make the above objectives, features and advantages of the present application more apparent, specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms, and should not be interpreted as being limited to the embodiments described herein, but rather, these embodiments are provided to more thoroughly and completely understand the present application. It should be understood that the drawings and embodiments of the present application are for illustrative purposes only, and are not intended to limit the scope of protection of the present application.

[0021] It should be understood that each step described in the method embodiments of the present application can be performed in different orders and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0022] The term "include," and variations thereof, as used in this document, is open-ended and means "to comprise but not to limit"; 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 additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optional" means "optional in at least one embodiment." Related terms shall be construed accordingly. It is to be noted that the terms "first," "second," and the like used in this document do not necessarily have an ordinal meaning with respect to one another, and are used to distinguish one element from another.

[0023] It should be noted that the terms "one" and "a" used in this document are intended to mean "one or more" unless otherwise indicated. It should be noted that the terms "first," "second," and the like used in this document are intended to distinguish between different elements, and are not intended to indicate an order of importance or a sequence of operations.

[0024] The names of the messages or information exchanged between the devices in the embodiments of the present application are used only for illustrative purposes and are not intended to limit the scope of the messages or information.

[0025] In combination Figure 1 As shown in the drawings, the laser cleaning path planning method provided by the embodiments of the present application comprises: The surface of the workpiece to be cleaned is imaged and collected, and coordinate registration and binarization processing are performed to obtain a binarized image of the surface of the workpiece to be cleaned.

[0026] Specifically, first, the workpiece to be cleaned, such as an aircraft workpiece with irregular paint layer residues, is adjusted to a fixed state by a tooling fixture, ensuring that the workpiece surface has no displacement deviation, providing a stable reference for subsequent operations. Subsequently, a pattern acquisition module coaxially assembled with the paint removal laser is started, and the pattern acquisition module performs a digital image acquisition on the base area covered by the laser scanning site of the surface of the workpiece to be cleaned; the matching of the image coordinates and the actual spatial coordinates of the workpiece is completed synchronously during the acquisition process, and the correspondence between the image pixel points and the physical positions of the workpiece surface is determined. In an optional embodiment of the present application, the scanning unit size is set to be consistent with the pixel unit size, i.e., the discrete coordinate system of the path point is equivalent to the image pixel coordinate system, realizing coordinate registration.

[0027] After the coordinate registration is completed, the original image collected is subjected to gray scale processing, filtering and denoising, and then a binary image is generated through adaptive threshold segmentation technology, the area of the dirt to be cleaned, such as the residual paint layer area on the surface of the aircraft, is marked as white, and the area of the workpiece substrate not to be cleaned, such as the damaged substrate exposed on the surface layer, is marked as black, so as to clearly distinguish the cleaning area from the non-cleaning area. Finally, the edge vertices of the binary image are calibrated, the spatial coordinate information of the image matrix is integrated, and the processed binary image data is uploaded to the local path heuristic module to provide a position basis for subsequent partitioning and path planning.

[0028] According to the binary image, the surface of the workpiece to be cleaned is regularly partitioned to obtain a plurality of sub-partitions, and a local path of each sub-partition is obtained through path covering.

[0029] Specifically, after the local path heuristic module receives the binary image data, the white cleaning area in the binary image is regularly partitioned by using the connected domain analysis method according to the geometric principle and the contour features of the cleaning area: the continuous white connected domain is divided into a plurality of independent and contour-regular sub-partitions, each of which can be individually planned for a path to ensure that the subsequent path covering is not missed or overlapped. At the same time, for each sub-partition, an inner spiral heuristic method is used to fill the local path, specifically, the geometric center of the sub-partition is taken as the starting point of the path, and the path is gradually extended outward along the spiral trajectory until the path covers the entire cleaning area of the sub-partition, so as to ensure the smoothness and cleaning efficiency of the local path and avoid resource waste caused by repeated paths. After the local path planning of all sub-partitions is completed, the local path information of each sub-partition is integrated into an indexable data packet and uploaded to the global ant colony adaptive path optimization module. In the embodiment of the present application, the local path information includes partition index, path starting / ending point coordinates and path trajectory data.

[0030] The starting points of the local paths of each sub-partition are mapped to obtain a global representative point set of all the local paths, and a traveling salesman problem model is generated according to the global representative point set.

[0031] Specifically, after the global ant colony adaptive path optimization module receives the local path data packet, it first analyzes the local paths of each sub-partition, and defines the starting point of the local path in each sub-partition as a global planning relay node, which represents the core position of the corresponding sub-partition in the global coordinate system of the workpiece. Then, through coordinate mapping technology, all relay nodes of the sub-partitions are converted into global representative points, and all global representative points are integrated to form a global representative point set, wherein each global representative point uniquely corresponds to a sub-partition.

[0032] Based on the spatial distribution characteristics of the global representative point set, the embodiment converts the sub-partition global access order optimization problem into a Traveling Salesman Problem (TSP) model, that is, to find a path that can traverse all global representative points, and to meet the shortest total path length, that is, to reduce the cleaning time and resource consumption, and each global representative point is traversed only once to avoid repeated cleaning. Through the conversion, the global path sorting problem between sub-partitions is solved, and the path cost degradation problem caused by parameter sensitivity in the path sorting of the traditional ant colony algorithm is avoided.

[0033] The Traveling Salesman Problem model is solved by the native heterogeneous competition-cooperation ant colony optimization algorithm to obtain the optimal global access sequence of all local paths.

[0034] Specifically, the native heterogeneous competition-cooperation ant colony optimization algorithm is used to solve the TSP model, and the process is divided into three stages.

[0035] Specifically, the first stage is the native heterogeneous competition stage, a control threshold (such as 60 observations) and a grouping step coefficient (such as 1 group every 3 observations, a total of 20 groups) are set, an incremental parameter mapping table is constructed within a preset parameter range (such as 0.01-20), and different parameters (which affect the search strategy of the ant colony) are assigned to the 20 groups. In this stage, the positive feedback update and iterative coverage mechanism in the ant colony algorithm are temporarily inhibited, and only the parameter adjustment is used to observe the path search performance of the ant colony: when the parameter is too large, the ant colony tends to choose the path according to fixed experience, and is easy to fall into local optimum; when the parameter is too small, the randomness of the ant colony search is enhanced, the path length is increased, and the result fluctuates obviously. At the same time, the path search result of the ant colony is analyzed for the change trend, and the change state of the path search is divided into three intervals: the oscillation zone, the rapid convergence zone, and the stable zone. Among them, the path length in the oscillation zone fluctuates violently, the path length in the rapid convergence zone is rapidly shortened, and the path length in the stable zone tends to be stable. The front interval of the stable zone corresponds to the parameter range with better search effect of the ant colony, which provides a basis for subsequent super parameter extraction.

[0036] The second stage is the super parameter extraction stage, and the support vector machine (FFD-SVM) model fused with the feedforward discrete passband diagnosis is used to analyze the path search data and parameter characteristics obtained in the above competition stage: the support vector machine is used to classify and predict the parameter characteristics, and the effective parameters are selected; combined with the feedforward discrete passband diagnosis technology, the prediction result is further optimized, the classification error is reduced, and the super parameter that can make the search effect of the ant colony optimal is accurately located, such as the optimal parameter value finally determined. This stage can realize high-precision parameter extraction under the condition of small sample, and ensure the optimization efficiency of the subsequent ant colony algorithm.

[0037] In the third stage, the ant colony cooperation pathfinding stage, the pheromone positive feedback updating mechanism in the ant colony algorithm is activated to extract the optimal hyperparameters to regulate the global search behavior of the ant colony. When the ant colony traverses the global representative points, it releases pheromones on the path it has passed through. Subsequent ants tend to choose paths with high pheromone concentrations. At the same time, pheromones gradually evaporate during the iteration process to avoid the path from falling into a local optimum. The total number of iterations (such as 300 times, including both the competitive stage observation and the cooperative stage iteration) is set. After multiple iterations, the ant colony converges to the optimal path solution, and the optimal traversal order of all sub-partition global representative points is obtained, that is, the optimal global access sequence of each sub-partition local path.

[0038] According to the optimal global access sequence, a laser cleaning scanning path of the surface of the workpiece to be cleaned is generated.

[0039] Specifically, the global ant colony adaptive path optimization module fuses the optimal global access sequence obtained with each sub-partition local path generated by the local path heuristic module, and connects each sub-partition in turn according to the order of the optimal global access sequence. At the junction of adjacent sub-partitions, a smooth transition technique (such as a curve transition) is used to eliminate path inflection points, ensuring the continuity of the laser motion trajectory and avoiding the decline in cleaning accuracy caused by frequent start and stop of the laser head. After fusion, the module integrates the final laser cleaning scanning path data into a data packet and uploads it to the subsequent laser execution control unit. The laser cleaning scanning path data includes sub-partition access order, local path trajectory coordinates, sub-partition junction transition path coordinates, etc. The laser execution control unit drives the laser head to perform cleaning work on the surface of the workpiece to be cleaned according to the path data, achieving efficient and precise laser intelligent cleaning.

[0040] The laser cleaning path planning method of the embodiment can accurately obtain the distribution information of the contaminants on the workpiece surface by imaging and collecting the surface of the workpiece to be cleaned and performing coordinate registration and binarization processing. The coordinate registration ensures the accurate correspondence between the image and the actual position of the workpiece surface, and the binarization processing simplifies the image into contaminant regions and non-contaminant regions, highlighting the key areas for cleaning and providing a clear basis for subsequent zoning and path planning. According to the binarized image, the workpiece surface is regularly zoned to obtain a plurality of sub-zones, and a local path is generated for each sub-zone. Dividing the workpiece surface into a plurality of sub-zones helps to decompose the complex cleaning task into a plurality of relatively simple sub-tasks. This decomposition approach can reduce the complexity of the problem, and generating a local path for each sub-zone can ensure that the contaminants in each region are effectively cleaned. For example, for a workpiece surface with irregular shape and uneven contaminant distribution, regular zoning can divide it into a plurality of small regions that are easy to handle. The local path in each small region can be designed flexibly according to the contamination situation in that region, improving the pertinence and efficiency of cleaning. Then, the starting points of the local paths are mapped to global representative points, the problem is transformed into a traveling salesman problem, and the traveling salesman solving method is used to find the optimal global access sequence. The present application makes full use of the optimization theory and algorithm of the traveling salesman problem to provide an effective framework for solving the laser cleaning path planning problem.

[0041] The traveling salesman problem model is solved by the original heterogeneous competition-cooperation ant colony optimization algorithm to obtain the optimal global access sequence of all local paths. Compared with the traditional ant colony algorithm, the original heterogeneous competition-cooperation ant colony optimization algorithm reduces the over-reliance on hyperparameters by introducing a competition-cooperation mechanism. During the algorithm running process, different ant colonies can compete and cooperate with each other, sharing information, so as to converge to the optimal solution more quickly. This algorithm can quickly re-plan the path under dynamic working conditions such as changes in the distribution of contaminants on the workpiece surface or adjustments in cleaning precision requirements, improving the sensitivity of laser cleaning path planning. For example, when new contaminated areas appear on the workpiece surface or the contamination level of certain areas increases, the algorithm can adjust the global access sequence in a timely manner to ensure that the cleaning path can adapt to the new situation, thereby improving the cleaning efficiency. According to the optimal global access sequence, a laser cleaning scanning path for the workpiece surface to be cleaned is generated. The finally generated laser cleaning scanning path considers the local paths of each sub-zone and the global optimal access sequence, and can achieve efficient and accurate laser cleaning. This path planning approach can ensure that the laser cleaning equipment accesses each contaminated area in the optimal order during the cleaning process, reducing the invalid movement of the equipment and improving the cleaning efficiency, solving the problem of low sensitivity of traditional ant colony algorithm path planning and difficulty in adapting to dynamic working conditions, and effectively promoting the large-scale application of laser intelligent cleaning technology.

[0042] In summary, the embodiment effectively solves the shortcomings of the traditional ant colony algorithm in dynamic path planning, greatly improves the sensitivity of path planning, and enables the entire laser cleaning path planning method to efficiently adapt to different working conditions, ultimately generating a high-quality laser cleaning scanning path.

[0043] Optionally, the imaging collection and coordinate registration and binarization processing of the surface of the workpiece to be cleaned are performed to obtain a binarized image of the surface of the workpiece to be cleaned, comprising: digitally imaging the surface of the workpiece to be cleaned to obtain an original gray-scale image of the surface of the workpiece to be cleaned; performing registration of the original gray-scale image and the actual spatial position of the surface of the workpiece to be cleaned through a coordinate conversion algorithm to obtain a registered gray-scale image; performing adaptive threshold segmentation or maximum inter-class variance segmentation on the registered gray-scale image to obtain the binarized image of the surface of the workpiece to be cleaned.

[0044] Specifically, select the on-demand maintenance workpiece with random irregularly shaped dirt, such as residual paint layer on the surface of an aircraft and industrial oil stain covered area, adjust it to a horizontal fixed state through a high-precision tool clamp, ensure that the workpiece surface does not shift or shake during the subsequent imaging process, and avoid deviation between image collection and actual position caused by workpiece offset. In a preferred embodiment of the present application, the imaging parameter calibration of the module can be performed by starting the image acquisition module coaxially assembled with the paint removal laser, the matching ratio of the imaging resolution and the laser scanning field is set, the pixel unit size of the collected image is equivalent to the laser scanning unit size, that is, the discrete coordinate system of the path point and the image pixel coordinate system are completely consistent in the subsequent path planning; at the same time, the exposure time and focal length of the module are calibrated to ensure that the collected image is clear and the dirt area and substrate area of the workpiece surface can be accurately distinguished. By controlling the calibrated image acquisition module, the laser scanning field corresponding to the substrate area of the workpiece surface to be cleaned, i.e. the actual covered area in the subsequent laser cleaning operation, performs a complete digital imaging collection, and obtains an original color image containing the dirt distribution and substrate contour details of the workpiece surface. The collected original color image is preprocessed by gray-scale conversion, the color image is converted into an original gray-scale image by using the weighted average method (such as calculating the gray-scale value according to the RGB channel weight 0.299, 0.587, 0.114), the brightness difference characteristics of the dirt and substrate in the image are retained in this process, the color interference is removed, the data dimension is simplified for subsequent coordinate registration and threshold segmentation, and the Gaussian filter algorithm is used for denoising processing of the original gray-scale image to eliminate the speckles caused by environmental light fluctuation and sensor noise, and ensure that the edge contour of the dirt and substrate in the image is clear and distinguishable.

[0045] Then, on the fixed surface of the workpiece to be cleaned, several characteristic edge vertices are selected as coordinate reference points, such as the corner vertices of the workpiece base and the boundary characteristic vertices of the dirt area. A high-precision displacement sensor is used to measure the three-dimensional coordinates of these reference points in the actual spatial coordinate system of the workpiece, i.e. the physical position coordinates, and the corresponding pixel coordinates of these reference points in the original gray-scale image are located and recorded, i.e. the two-dimensional coordinates in the image coordinate system. A perspective transformation coordinate conversion algorithm is used to establish a one-to-one mapping relationship between the pixel coordinate system of the original gray-scale image and the actual spatial coordinate system of the surface of the workpiece to be cleaned, based on the corresponding pairs of actual spatial coordinates and pixel coordinates of the calibrated reference points. Through this mapping relationship, the coordinates of any pixel point in the original gray-scale image are directly converted into the actual physical position coordinates of the workpiece surface, realizing accurate registration of the original gray-scale image and the actual spatial position of the workpiece, and obtaining the registered gray-scale image. This registration ensures that the subsequent image-based path planning coordinates can be directly applied to the laser cleaning execution unit, avoiding path deviation.

[0046] The registered gray-scale image is again subjected to smoothing processing (such as median filtering) to further eliminate the influence of residual noise on the segmentation effect, ensuring that the gray-scale difference between the dirt area to be cleaned and the non-cleaning base area in the image is more significant, laying a foundation for accurate segmentation. If there is local uneven illumination on the surface of the workpiece to be cleaned, such as uneven gray-scale distribution caused by partial shading, an adaptive threshold segmentation algorithm is further used. The algorithm dynamically adjusts the segmentation threshold according to the gray-scale mean of different local areas of the image, ensuring that the dirt and the base can be accurately distinguished in each local area. For example, the segmentation threshold is lowered in darkly illuminated areas and raised in brightly illuminated areas, avoiding partial dirt area missing or misclassification due to illumination differences. If the illumination conditions of the surface of the workpiece to be cleaned are uniform, and the gray-scale difference between the dirt area and the base area is stable overall, such as the overall dirt gray-scale being higher or lower than the base gray-scale, the Otsu algorithm is used. The algorithm calculates the inter-class variance of the dirt pixel class and the base pixel class under each possible gray-scale threshold by traversing all possible gray-scale thresholds, and selects the threshold with the maximum inter-class variance as the optimal segmentation threshold, thereby realizing accurate segmentation in the global range.

[0047] After segmentation, the area in the image determined to be the dirt to be cleaned is marked as white, such as setting the pixel value to 255, and the area determined to be the base that does not need to be cleaned is marked as black, such as setting the pixel value to 0, generating a binary image of the surface of the workpiece to be cleaned. Then, the edge profile of the binary image is re-calibrated, and the spatial coordinate information corresponding to the image matrix is integrated. The binary image data is uploaded to the local path heuristic module of the laser paint removal intelligent planning system in the form of indexable data packets, providing clear regional position basis for subsequent regular partitioning and local path planning of the area to be cleaned.

[0048] In this optional embodiment, by directional digital imaging of the workpiece surface and converting to the original gray image, the brightness difference characteristics of the dirt area and the base area can be effectively preserved, and environmental interference can be eliminated through denoising processing, laying a data foundation for subsequent accurate differentiation of the to-be-cleaned area. Secondly, the registration of the original gray image and the actual spatial position of the workpiece is realized by means of coordinate conversion algorithm, and a one-to-one mapping relationship between image pixel coordinates and workpiece physical coordinates is established by setting the pixel unit size equivalent to the laser scanning unit size, eliminating the deviation between the image and the actual position, and ensuring that the path coordinates based on image planning can be directly and accurately mapped to the workpiece surface, avoiding missed cleaning or false cleaning caused by coordinate misplacement during laser cleaning. Finally, according to the illumination conditions of the workpiece surface, the adaptive threshold segmentation is flexibly selected to cope with the local uneven illumination scene or the maximum inter-class variance segmentation to cope with the stable global gray difference scene, which can realize efficient and accurate differentiation of the to-be-cleaned dirt area and the non-cleaning base area, and the generated binary image can clearly and intuitively present the outline and position of the to-be-cleaned area, providing accurate and reliable area position basis for the subsequent rule-based partitioning, path filling of the local path inspiration module and global planning of the global ant colony adaptive path optimization module, thereby effectively ensuring the accuracy and effectiveness of the laser cleaning path planning, and providing key technical support for reducing the time cost, resource consumption of the cleaning process and improving the comprehensive efficiency of laser intelligent cleaning.

[0049] Optionally, the rule-based partitioning of the to-be-cleaned workpiece surface according to the binary image comprises: Performing connected component analysis on the foreground pixels in the binary image to obtain a plurality of to-be-cleaned connected domains; For each of the to-be-cleaned connected domains, extracting the minimum circumscribed contour region of the to-be-cleaned connected domain; According to the field lens scanning range of the laser cleaning device, the cleaning accuracy requirement and the path planning efficiency demand, a partition size threshold is determined; According to the partition size threshold, the to-be-cleaned connected domains are adaptively divided to obtain a plurality of sub-partitions.

[0050] Specifically, after the local path heuristic module receives the binarized image data, firstly, the white foreground pixels in the image, i.e. the to-be-cleaned dirt area, are subjected to connected component analysis, and an eight-neighbor connectedness judgment algorithm is adopted to traverse all the pixel points of the binarized image, and the adjacent (up, down, left, right, and diagonal adjacent) white foreground pixels are classified into the same connected unit, thereby separating all the independent to-be-cleaned connected domains in the image, such as the multiple independent residual paint layer areas scattered on the surface of the workpiece in the aircraft paint layer cleaning scene, each of which forms a to-be-cleaned connected domain; meanwhile, the pixel coordinate range and the pixel number of each to-be-cleaned connected domain are recorded to clearly determine the spatial distribution and the size of each connected domain. For each separated to-be-cleaned connected domain, the local path heuristic module extracts the edge features of the connected domain through a contour detection algorithm (such as Canny edge detection combined with contour fitting), and then generates the minimum circumscribed contour region of the connected domain based on the geometric fitting principle. If the connected domain is in an irregular shape (such as an arc-shaped or polygonal residual paint layer), the minimum circumscribed contour region that can completely wrap the connected domain and has a regular contour is generated, which is preferably a rectangle in a preferred embodiment of the present application, and can be a polygon in special cases, so as to ensure that the circumscribed contour region only contains the to-be-cleaned connected domain and a small amount of substrate edge pixels, thereby avoiding missing the to-be-cleaned area and reducing the inclusion of invalid substrate area. Meanwhile, the vertex coordinates of each minimum circumscribed contour region are recorded, and in the present embodiment, the vertex coordinates mainly refer to the nine-point calibration in the camera coordinate calibration process, including the vertices of the Chinese character and the midpoint, so as to realize the alignment of the pixel coordinate system and the scanning coordinate system, and further provide substantial matching coordinate basis for the focal field scanning of the laser, thereby forming regularized regional boundary data.

[0051] By calling the preset parameter configuration interface through the local path heuristic module, the partition size threshold is determined in combination with the core performance parameters of the laser cleaning device and the actual operation requirements. Specifically, the field lens scanning range of the laser cleaning device is taken as the basis for constraint, such as the maximum transverse / longitudinal size of single effective scanning of the field lens, so as to ensure that the maximum size of a single sub-partition does not exceed the single scanning coverage range of the field lens, thereby avoiding the path connection error caused by the cross-field lens range. In combination with the cleaning accuracy requirement, such as the micron-level cleaning accuracy scene, the lower limit of the partition size is set, such as the minimum edge length of the partition being not less than 5 times the laser spot diameter, so as to prevent the partition from being too small to cause the local path to be too fragmented, thereby affecting the cleaning continuity and accuracy. At the same time, the path planning efficiency requirement is taken into account, and the threshold is adjusted to the balance interval of the single-partition path filling time consumption and the partition connection time consumption through historical data calibration, such as the cleaning efficiency test of the same workpiece type, so as to avoid the increase of the global path sorting complexity caused by excessive partitions. Finally, the determined partition size threshold is stored in the form of maximum edge length x minimum edge length data, which serves as the judgment standard for subsequent adaptive division.

[0052] Finally, the minimum circumscribed contour region size of each to-be-cleaned connected domain is compared with the determined partition size threshold by the local path heuristic module to perform adaptive partitioning. Specifically, if the minimum circumscribed contour region size of the to-be-cleaned connected domain is less than or equal to the partition size threshold, such as the circumscribed rectangle side length of a small independent residual paint layer, the minimum circumscribed contour region is directly taken as an independent sub-partition without additional segmentation; if the minimum circumscribed contour region size of the to-be-cleaned connected domain is greater than the partition size threshold, such as the circumscribed rectangle side length of a large continuous residual paint layer, the circumscribed contour region is partitioned based on the partition size threshold by using the grid equal division + edge adaptation manner, the circumscribed contour region is first divided into several standard sub-partitions with a size equal to the threshold by constructing a virtual grid according to the threshold size, and then the non-standard size region remaining at the edge, which is less than the threshold but greater than the size lower limit, is separately partitioned into one adaptive sub-partition, so that the size of all sub-partitions is within the range of the partition size threshold; After the partitioning is completed, a unique index number is assigned to each sub-partition, the vertex coordinates and the original connected domain identifier of each sub-partition are recorded to form a regularized sub-partition data set, and a structured region unit is provided for subsequent sub-partition local path filling and global ant colony adaptive path optimization.

[0053] In this optional embodiment, the foreground pixels in the binary image are subjected to connected domain analysis to accurately separate multiple independent to-be-cleaned regions on the workpiece surface, such as residual paint layers and oil stains at different positions, so as to avoid mutual interference of different to-be-cleaned regions in subsequent partitioning and path planning. At the same time, the minimum circumscribed contour region of each to-be-cleaned connected domain is extracted to convert the irregularly shaped dirt region into a boundary regular region unit, eliminate the influence of shape disorder on the partitioning specification, and provide a unified region reference for subsequent standardized partitioning. The partition size threshold is determined in combination with the field lens scanning range of the laser cleaning device, the cleaning accuracy requirement, and the path planning efficiency requirement, so that the threshold has both equipment adaptability and operation practicability. Adaptive partitioning of the connected domain based on the threshold can ensure that the size of all sub-partitions is reasonable, avoiding both the increase in path planning difficulty caused by excessively large partitioning and the frequent path connection problem caused by excessively small partitioning. The finally generated regular sub-partition not only provides ordered and controllable region units for subsequent sub-partition local path filling, but also lays a structured foundation for sub-partition indexing and global path planning of the global ant colony adaptive path optimization module, effectively improving the specification, accuracy, and efficiency of laser cleaning path planning, ensuring the continuity and accuracy of subsequent cleaning operation, and reducing the planning complexity and time cost of intelligent cleaning process.

[0054] Optionally, the path covering for each of the sub-partitions to obtain the local path of each of the sub-partitions comprises: determine a scanning reference direction of the local path according to the shape of the sub-partition and the scanning direction adaptability of the laser cleaning device; obtain a starting point, an ending point and a path node sequence in the sub-partition according to the scanning reference direction and a path covering algorithm corresponding to the shape of the sub-partition; generate an initial local path of each sub-partition according to the starting point, the ending point and the path node sequence in the sub-partition; optimize the initial local path by a Bezier curve to obtain the local path.

[0055] Specifically, the shape data of each sub-partition is read by the local path heuristic module, wherein the shape data such as a regular contour like a rectangle, a polygon, and a coordinate range are combined with the scanning direction characteristics of the laser cleaning device, such as the motion smoothness of the laser head horizontal / vertical scanning and the difference in turning energy consumption, to determine the scanning reference direction of the local path. If the sub-partition is a rectangle, the direction parallel to the long side of the sub-partition is preferentially selected as the reference direction, so as to reduce the number of laser head turns and improve the scanning efficiency. If the sub-partition is an irregular polygon, the longest axis direction of the sub-partition is extracted by geometric analysis and is set as the reference direction, so as to ensure the continuity of the laser motion trajectory in the path covering process. At the same time, the reference direction conflicting with the limit direction of the laser head motion is avoided, such as avoiding frequent triggering of the laser head motion stroke boundary, so as to ensure that the reference direction is adapted to the shape of the sub-partition and meets the motion characteristics of the laser device.

[0056] Based on the determined scanning reference direction, the local path heuristic module calls a preset path covering algorithm. In an optional embodiment, an inner spiral heuristic method is adopted to perform path covering calculation for each sub-partition, taking the geometric center of the sub-partition as the path starting point, ensuring coverage from the center to the edge of the sub-partition, reducing the risk of edge washing, and extending the path outward along the scanning reference direction at a preset step length, wherein the preset step length is equal to the laser spot diameter, ensuring gapless coverage. In the extending process, the boundary intersection points of the path and the sub-partition contour are detected in real time, the intersection point finally reaching the edge of the sub-partition is set as the ending point, and the coordinates of all turning points and boundary intersection points in the path extending process are recorded to form a path node sequence containing the starting point, the ending point and the intermediate turning points. If the shape of the sub-partition is special, such as having an inner recess contour, the non-cleaning part of the inner recess area is automatically avoided when the path node sequence is generated, so as to ensure that the node sequence only covers the cleaning area of the sub-partition.

[0057] The local path heuristic module connects all the nodes in the path node sequence in a straight line in the order of the starting point, the intermediate turning point and the ending point to form an initial local path for each sub-partition; meanwhile, in the connection process, the path is checked in real time to see whether it is completely located within the sub-partition range, and if the path exceeds the outline of the sub-partition, such as the edge node calculation deviation, the coordinates of the corresponding node are automatically adjusted to the boundary of the sub-partition to ensure that the initial local path completely covers the cleaning area of the sub-partition without redundant paths, thereby providing a basic path framework for subsequent path smoothing optimization.

[0058] In this embodiment, to eliminate the inflection points formed by the straight-line connection between the nodes in the initial local path and avoid the decrease in cleaning accuracy and the increase in energy consumption caused by the frequent sharp turning of the laser head, the local path heuristic module performs transition optimization on the initial local path by using a Bezier curve. Specifically, for two adjacent nodes in the initial path, especially the turning point and the boundary intersection point, the tangent direction between the nodes is calculated, and a third-order Bezier curve is constructed as a transition segment. By adjusting the control point parameters of the Bezier curve, the transition segment is smoothly connected with the straight-line segment of the adjacent node to ensure that the optimized path has continuous curvature and no obvious inflection points. After optimization, the final local path of each sub-partition is generated, and the starting point coordinates, ending point coordinates, complete path trajectory data and sub-partition index of the local path are stored in association to be uploaded to the global ant colony adaptive path optimization module in the form of an indexable data packet, thereby providing data support for the subsequent concatenation of the global path and the precise execution of the laser cleaning action.

[0059] In this optional embodiment, by determining the scanning reference direction according to the adaptability of the sub-partition shape and the scanning direction of the laser cleaning device, the local path planning can be accurately matched with the motion characteristics of the laser head, the number of invalid turning of the laser head can be reduced, and the foundation for improving the scanning efficiency is laid. The starting point, the ending point and the path node sequence are obtained by combining the scanning reference direction and the path coverage algorithm corresponding to the sub-partition shape, which can ensure that the path nodes completely cover the cleaning area of the sub-partition and avoid missing cleaning or redundant paths caused by missing nodes. When the initial local path is generated based on the starting point, the ending point and the path node sequence, the deviation nodes are adjusted by real-time checking whether the path is located within the sub-partition range, which can further guarantee the integrity and accuracy of the initial path and exclude invalid paths that exceed the sub-partition. Finally, the initial local path is transitionally optimized by a Bezier curve to eliminate the inflection points formed by the straight-line connection between the nodes, so that the curvature of the optimized local path is continuous, the decrease in cleaning accuracy and the increase in energy consumption caused by the frequent sharp turning of the laser head are effectively avoided, the smooth local path foundation is provided for the concatenation of the sub-partition path by the subsequent global ant colony adaptive path optimization module, the continuity, accuracy and energy efficiency of the laser cleaning operation are improved, and the path planning error and resource consumption in the intelligent cleaning process are reduced.

[0060] Optionally, the starting point of each local path of each sub-partition is mapped to obtain a global representative point set of all local paths, comprising: Obtaining the local coordinates of the starting point of each sub-partition in the local coordinate system of the sub-partition; According to the local coordinates of the starting point, determining the coordinate registration relationship between the image pixels and the actual spatial position of the sub-partition; According to the coordinate registration relationship, converting the local coordinates of the starting point into global coordinates in the global coordinate system; The global coordinates of the starting point of all sub-partitions are summarized to form the global representative point set.

[0061] Specifically, the global ant colony adaptive path optimization module calls the sub-partition data packet uploaded by the local path heuristic module to extract the starting point information of each sub-partition local path, which is the geometric center of the sub-partition determined in the local path planning stage or the optimal starting position adapted to the shape of the sub-partition, and carries the local coordinates of the starting point in the sub-partition local coordinate system. The sub-partition local coordinate system takes the left lower corner vertex of the minimum circumscribed contour as the origin, takes the axis parallel to the scanning reference direction of the sub-partition as the horizontal and vertical axes, and the unit is consistent with the image pixel coordinate system. The module reads and stores the local coordinates of the starting point one by one according to the sub-partition index order, such as the unique number assigned when the sub-partition is divided, to form a corresponding list of sub-partition index-starting point local coordinates, ensuring that each sub-partition has a clear local position identifier.

[0062] The global ant colony adaptive path optimization module calls two types of coordinate correlation relationships established by the image acquisition module in the imaging acquisition stage of the surface of the workpiece to be cleaned from the coordinate database of the system. One is the registration relationship between the image pixel coordinate system and the actual spatial coordinate system of the workpiece to be cleaned. Specifically, based on the perspective transformation algorithm, the size mapping ratio between the pixel unit and the actual space unit is constructed. By setting the pixel unit size equivalent to the laser scanning unit size, the ratio is 1:1, that is, 1 pixel unit corresponds to 1 actual scanning unit. The second is the correlation relationship between the sub-partition local coordinate system and the image pixel coordinate system, that is, the pixel coordinates of the origin of the sub-partition local coordinate system in the image pixel coordinate system, and the direction angle between the axis of the local coordinate system and the axis of the pixel coordinate system. Since the sub-partition is a regular partition, the direction angle is 0°, and only the origin pixel coordinates need to be called. The module integrates the two types of relationships into a complete mapping link of sub-partition local coordinates-image pixel coordinates-workpiece actual spatial coordinates, which serves as the basis for subsequent coordinate conversion calculation, ensuring that the conversion process is unbiased.

[0063] Based on the coordinate registration relationship, the global ant colony adaptive path optimization module performs two-step conversion on the local coordinates of each sub-partition starting point. Specifically, in the first step, the local coordinates of the starting point are converted into image pixel coordinates. By superimposing the local coordinates of the starting point on the pixel coordinates of the origin of the sub-partition local coordinate system, the pixel coordinates are obtained. It should be noted that, since the direction angle is 0°, no additional rotation calculation is required. In the second step, the image pixel coordinates are converted into global coordinates in the global coordinate system, i.e., the actual spatial coordinate system of the workpiece to be cleaned. Based on the 1:1 size mapping ratio between the pixel coordinate system and the actual spatial coordinate system, the pixel coordinates are directly taken as the global coordinates of the starting point. In an embodiment of the present application, if there is a slight calibration deviation in the global coordinates, a pre-set correction coefficient can be used for fine tuning to ensure complete consistency with the actual position of the workpiece. After the conversion is completed, the module records the global coordinates of the starting point corresponding to each sub-partition index, forming the associated data of sub-partition index-starting point global coordinates. The global ant colony adaptive path optimization module collects and integrates the global coordinates of all sub-partition starting points in the order of sub-partition indexes. The global coordinates of each starting point are taken as a global representative point, which uniquely corresponds to a sub-partition and can represent the core position of the sub-partition in the actual space of the workpiece. Each global representative point is assigned an index identifier consistent with the sub-partition, and finally a structured data set containing global representative point coordinates-sub-partition indexes is formed, i.e., a global representative point set. In an embodiment of the present application, the representative point is taken as each local sub-partition, i.e., each independent sub-partition is regarded as a serialized sub-node under global planning. In spatial distribution, the point closest to the center of the whole graph of each sub-partition can be taken to represent the spatial coordinates of the sub-partition. The spatial domain distribution characteristic is to represent the distribution characteristics of the sub-partition and the adaptive (calculation reference) characteristics in the global path optimization process by using spatial coordinates and spatial Euclidean distance. This data set serves as input data for subsequent construction of a Traveling Salesman Problem (TSP) model, providing basic location information for solving the global access sequence of the sub-partition.

[0064] In the optional embodiment, by obtaining the local coordinates of each sub-partition local path starting point in the sub-partition local coordinate system, the relative position reference of the starting point in the sub-partition can be determined, avoiding the confusion of subsequent coordinate conversion caused by ambiguous starting point position identification, and providing a clear and unified local position basis for subsequent global coordinate mapping. At the same time, the coordinate registration relationship between the sub-partition image pixels and the actual spatial position is determined by combining the starting point local coordinates, which can accurately establish the correlation link between the sub-partition local coordinate system and the actual spatial coordinate system of the workpiece, eliminate the mapping fault between the local coordinates and the global coordinates, ensure that the bottom logic of coordinate conversion is consistent with the technical parameters of the previous imaging and partitioning link, and avoid introducing additional errors due to registration relationship deviation. According to the coordinate registration relationship, the starting point local coordinates are converted into global coordinates in the global coordinate system, which can accurately map the starting point representing only the relative position in the sub-partition to the physical position in the actual space of the workpiece, completely solve the problem of disconnection between the local path coordinates and the actual cleaning position, and ensure that the subsequent path planning based on global coordinates can directly match the actual cleaning requirements of the workpiece, preventing path deviation, missed cleaning or false cleaning during laser cleaning. Finally, the global coordinates of all sub-partition starting points are summarized to form a global representative point set. This set converts the dispersed sub-partitions into a unified unit recognizable by global path optimization in the form of global representative point-sub-partition correspondence, provides input data for the subsequent conversion of the sub-partition global access sequence optimization problem into a Traveling Salesman Problem (TSP) model, and makes the global ant colony adaptive optimization algorithm have a clear optimization object, significantly improving the orderliness and efficiency of global path planning. At the same time, it lays the foundation for subsequent global access sequence solving and final laser cleaning path generation, further ensuring the comprehensive precision and efficiency of laser intelligent cleaning.

[0065] Optionally, the generating a Traveling Salesman Problem model according to the global representative point set comprises: taking each of the starting points in the global representative point set as a node in the Traveling Salesman Problem model, and calculating the Euclidean distance between any two nodes; taking the Euclidean distance as the path cost between the two nodes, traversing all the nodes to determine the total path cost; taking the minimum total path cost as the optimization objective, and constructing the Traveling Salesman Problem model comprising a node set, a path cost matrix, and a traversal constraint condition.

[0066] Specifically, the global ant colony adaptive path optimization module reads the global representative point set data, and defines the starting point of each sub-partition local path in the set as a node to be traversed in the traveling salesman problem model. Each node is uniquely associated with a sub-partition, and the corresponding sub-partition number and local path information can be traced through the node index, ensuring that the model nodes match the sub-partitions to be covered by laser cleaning one by one, without missing or redundant nodes. For any two nodes to be traversed in the model, the spatial straight line distance calculation logic is used to traverse all node combinations and calculate the Euclidean distance between them. The calculated Euclidean distance is directly mapped as the path cost between the two nodes. According to the rule that all nodes are traversed and each node is visited only once, for any node traversal order (such as node1→node2→…→node n→node1), the path cost between adjacent nodes in the traversal process is sequentially added to obtain the total path cost corresponding to the traversal order, and the core quantitative indicator of the model to be optimized is determined. At the same time, taking the minimum total path cost as the core optimization objective, all nodes to be traversed are collected to form a node set, each node in the set is attached with global coordinates and sub-partition index, and the correspondence between the nodes and the sub-partitions is ensured; the path cost between all nodes such as node i-node j is arranged in the form of an n×n two-dimensional matrix according to the node index order, where n is the number of nodes, and the matrix element C(i,j) represents the path cost from node i to node j. In this embodiment, the matrix form is used to facilitate the subsequent fast calling and calculation of the native heterogeneous competition-cooperation ant colony optimization algorithm, thereby improving the algorithm solving efficiency. At the same time, each node is set to be visited at least once and at most once, which ensures that all sub-partitions are covered without missing cleaning or repeated cleaning, the path is a continuous traversal sequence, and the constraints of avoiding meaningless empty movement of the laser head between non-adjacent nodes and reducing resource waste are avoided, thereby ensuring that the model output meets the requirements of path integrity and continuity for actual laser cleaning operations.

[0067] In this optional embodiment, by defining the starting point of the corresponding sub-partition in the global representative point set as a node in the traveling salesman problem model, the model nodes can be accurately matched with the actual cleaning sub-partitions, avoiding the problem of disconnection between the subsequent path planning and the actual position of the workpiece caused by node position deviation, and ensuring that the cleaning operation is complete without missing or misplacing. Using the Euclidean distance between any two nodes as the path cost can directly associate the abstract cost with the actual path length, time and energy consumption of the laser head movement, so that the optimization objective of minimizing the total path cost can truly reflect the demand for reducing cleaning resource consumption. By constructing the model with the constraints of each node being visited at least once and at most once and the path being continuously traversed, all sub-partitions can be fully covered without repeated access, and invalid movement can be avoided. At the same time, the standardized node set and path cost matrix can adapt to the fast solving of the subsequent native heterogeneous competition-cooperation ant colony optimization algorithm, reduce the time-consuming of algorithm parameter conversion, and improve the path planning efficiency and accuracy.

[0068] Optionally, the step of solving the Traveling Salesman Problem model using the native heterogeneous competition-cooperative ant colony optimization algorithm to obtain the optimal global access sequence for all local paths includes: Based on the ant population size, initial pheromone concentration, pheromone volatile factor, and preset regulatory threshold and preset grouping step size coefficient of the original heterogeneous competition stage, a parameter mapping table of the expected heuristic factor is constructed. In native heterogeneous competition, keeping the initial pheromone concentration unchanged, the expected heuristic factors are injected into the ant population in the order of the parameter mapping table, driving the ants to independently complete path construction, and obtaining the total path cost, the first-order difference and the second-order difference of the total path cost; When the number of iterations of the ant population reaches a preset control threshold, the second-order difference is classified by support vector machine to obtain the fluctuation convergence segment, and the expected heuristic factor corresponding to the smallest first-order difference in the fluctuation convergence segment is taken as the optimal expected heuristic factor. Based on the optimal expectation heuristic factor, and combined with the probability transition algorithm, the ant colony iterative process is used to obtain the optimal global access sequence.

[0069] Specifically, a parameter mapping table of desired heuristic factors is constructed based on the ant population size, initial pheromone concentration, pheromone evaporation factor, and preset control thresholds and preset grouping step size coefficients for the primary heterogeneous competition stage. Then, during the primary heterogeneous competition stage, keeping the initial pheromone concentration constant, the desired heuristic factors are injected into the ant population sequentially according to the order of the parameter mapping table, driving the ants to independently complete path construction. When the number of iterations in the ant population reaches the preset control threshold, a support vector machine is used to perform binary classification on the second-order differences, thereby obtaining the fluctuating convergence segment. Furthermore, the desired heuristic factor corresponding to the smallest first-order difference within the fluctuating convergence segment is taken as the optimal desired heuristic factor. Finally, based on the optimal desired heuristic factor, a probabilistic transition algorithm is used to iterate the ant colony, ultimately obtaining the optimal global access sequence.

[0070] In a preferred embodiment of the present invention, the specific implementation is as follows: After the local partitioning module has finished executing, a global sequence serialization taboo table is defined. , For the first Node and the The set of paths between nodes. It should be noted that consecutive nodes involved in the path optimization process of the ant colony algorithm are random nodes, represented here by d and q. In the ant colony optimization algorithm... In the next iteration, there are Probability of ant seeking the best at any given moment: (1); in, a candidate node set accessible to global ants in a single iteration, i.e., a sub-partition sequence set not yet visited in the routing process, , all selectable nodes, a pheromone heuristic factor in the global ant colony optimization process, a pheromone heuristic factor in the Xth iteration, where let denote the global dynamic optimal expected heuristic factor, a global heuristic function matrix associated with node d and node set q at time t, a global heuristic function matrix associated with node d and node set q at time t, a global heuristic function matrix associated with node d and node set q at time t, a global heuristic function matrix associated with node d and node set q at time t, is defined as where is the Euclidean distance between node d and node q, is the Euclidean distance between node d and node q, is the Euclidean distance between node d and node q, is the Euclidean distance between node d and node q, is the Euclidean distance between node d and node q, is the pheromone matrix associated with node d and node q at time t, is the pheromone matrix associated with node d and node set q at time t, is the pheromone matrix associated with node d and node q at time t, is the pheromone matrix associated with node d and node set q at time t, is the pheromone matrix associated with node d and node q at time t, is the pheromone matrix associated with node d and node set q at time t, is the pheromone matrix associated with node d and node q at time t, is the pheromone matrix associated with node d and node set q at time t, is the pheromone matrix associated with node d and node set q at time t, is the pheromone matrix associated with node d and node set q at time t, denotes the global heuristic function matrix associated with node d and node set q at time t.

[0071] take is the heterogeneous regulation boundary threshold value, then in the Xth primary heterogeneous regulation process, the dynamic mapping relationship of the global expected heuristic factor can be expressed as: (2); wherein, is a heuristic factor dynamic parameter index variable. Further, the pheromone update process of the global ant colony optimization can be expressed as (3); (4); (5); wherein w is the ant population size, is the initial pheromone concentration, is the pheromone concentration after the Xth iteration, the pheromone concentration on path d-q at time t, the pheromone increment for this period, the pheromone increment for the first the amount of pheromone released by the only ant on path d-q in the xth iteration, the pheromone increment for the first the objective function value of the only ant in the xth iteration, the global optimization pheromone evaporation factor, Q is the global optimization pheromone intensity, and Q is set to 1 here, also in view of its redundancy with the expected heuristic factor.

[0072] Decomposition of the number of iterations x can be obtained: , (6); For any iteration number x, the first order difference of the global optimization objective function between the second and third strokes is defined as And its second order difference function is expressed as: (7); (8); In order to realize high-precision prediction segmentation under small sample condition, the present application proposes a support vector machine prediction model FFD-SVM) which fuses feedforward discrete passband diagnosis, and its mathematical modeling is as follows: , (9); Wherein, is the second order difference gain amplitude classification output of SVM in the whole iteration period, is the binary classification label space, which is set to 1, -1, is the threshold parameter of the feedforward discrete passband filter.

[0073] (10); Wherein, is the passband reference bias operator, is the threshold gain coefficient when the sample size is , and the kernel function of SVM can be characterized as: (11); Wherein, are the training data matrix and label vector respectively, is the weight bandwidth parameter of the radial basis function, and are the structure parameters of the SVM model.

[0074] In the optimal hyperparameter optimization stage, the input variables of the category are initialized index counter satisfy: , (12); Through derivation, the dynamic parameter index variable : , (13); Further, the optimal expected heuristic factor mapped thereby is obtained: (14).

[0075] Optionally, the generating of the laser cleaning scanning path of the surface of the workpiece to be cleaned according to the optimal global access sequence comprises: mapping the node order in the optimal global access sequence to the execution order of the corresponding sub-partition of the node; extracting the local path of each sub-partition in turn according to the execution order of the sub-partition, and splicing the local paths to obtain a spliced overall path; speed planning is performed on the spliced overall path to generate the laser cleaning scanning path.

[0076] Specifically, the correspondence between the node sequence and the sub-partition is established, the node order in the optimal global access sequence is mapped to the execution order of the corresponding sub-partition, thereby providing clear guidance for subsequent path splicing. Through this mapping, the laser cleaning head can access each sub-partition according to the optimal global order, avoiding unnecessary back-and-forth and repeated operations, and significantly improving the cleaning efficiency. Next, the local path of each sub-partition is extracted in turn according to the execution order of the sub-partition, and spliced to form a spliced overall path. Finally, the spliced overall path is speed planned to generate the final laser cleaning scanning path. In the embodiment of the present application, the speed planning involves reasonably adjusting the moving speed of the laser head according to the specific requirements of the cleaning task, the performance parameters of the laser equipment, and the material properties of the workpiece. For example, when cleaning difficult-to-remove contaminants, the speed may need to be reduced to ensure cleaning effect; while cleaning large areas of lightly contaminated areas, the speed can be appropriately increased to improve efficiency. Through speed planning, not only can the cleaning quality be improved, but also the energy consumption and equipment wear and tear can be reduced.

[0077] ​This optional embodiment effectively solves the problems of path discontinuity, low efficiency, and inconsistent cleaning results in traditional path planning methods. Through precise node mapping, path splicing, and speed planning, a highly efficient scanning path suitable for laser cleaning is generated, significantly improving the practicality and reliability of laser intelligent cleaning technology.

[0078] In a preferred embodiment of the present invention, an application example of cleaning residual paint layers from an aircraft is used to further illustrate the present invention. The specific steps are as follows: Step one: Extract image features and coordinates of the target. First, fix the workpiece to be maintained. As described in the invention, use the image acquisition module to sample, map coordinates, and binarize the surface of the workpiece within the field of view, generating a binarized feature result containing spatial coordinate information, and generating the corresponding target feature binarized image, as shown below. Figure 2 As shown, the white area represents the topcoat area, and the black area represents the surface damage area.

[0079] Step two: Based on the required maintenance scheme, the binary feature map is topologically partitioned and the target region path is filled using the local path heuristic module. This example demonstrates a custom maintenance scheme for aircraft paint cleaning residues, which simplifies the topology of each connected component into sequentially indexable sub-intervals, and then partitions each sub-partition... Figure 2 The white area in the middle is used for scanning path coverage. In this example, the inner spiral heuristic is used to fill the path of the sub-partitions, and the data is uploaded to the global adaptive optimization module described in this invention in the form of a global representative node index (that is, each sub-partition is regarded as a global node with sequence information and coordinate information, and the representative coordinates of each sub-partition are explained as above). The partition and local path coverage diagram is shown below. Figure 3 As shown.

[0080] Step three uses the path inspiration within the sub-partition obtained in step two as a transition node for global-local mode switching, realizing continuous and smooth planning of the planar target coverage path, and demonstrating the ant colony algorithm's hyperparameter adaptive optimization process in global planning. Combined with... Figure 4 As shown, Figure 4 Presenting the observation results of parameter competition under the native heterogeneous regulation of ant colonies: By balancing the identifiability of the sampling model features with the overall system operation time, a regulation threshold is set. The number of trials is 60, with a grouping step size coefficient of 3 (as shown in Formula 6), meaning that every 3 observations of the same parameter are divided into 1 group, making the control process equally subdivided into 20 competing groups. An incremental β parameter mapping table is constructed within the value range of 0.01-20, and the competitive parameters are sequentially mapped to the 20 competing groups. At this time, the increase in the number of competitions is equivalent to the corresponding increase in the β parameter. According to Formulas 3-5, a failure operation is performed on the pheromone update mechanism and iterative optimization process during the control process.

[0081] In the native heterogeneous competition process, the embodiments of the present invention address... Figure 4 The trend analysis of the curve change is as follows: The change of the β parameter can be directly mapped to the expected heuristic weight, and its impact on the path search distance is as follows: Both excessively large and excessively small β values ​​are detrimental to path distance optimization. In the absence of pheromone intervention, when the β value is large, the ant colony tends to adopt a prior probability selection strategy, reducing randomness and significantly enhancing local optima; when the β value is small, it tends to favor random search, leading to an increase in path search distance and significant fluctuations in the iteration results within the group. Analysis of the intra-group fluctuation characteristics reveals that a significant step change usually occurs in the third observation within the initial competitive group. Figure 4 The first six groups of step characteristics and their directions of change are marked with dashed boxes and arrows. This phenomenon indicates a cross-correlation between the β value and the randomness of ant colony search. Based on Formula 2, forward differencing is performed on the third observation of each group, and the calculated change characteristic curve is shown below. Figure 5 As shown by the red dashed line, observations indicate that the fluctuation state of the first-order change in path search can be divided into three characteristic intervals as the β value increases: the oscillation region, the fast convergence region, and the stable region. The convergence region has a relatively short convergence time and can be approximated as the critical boundary. The early part of the stable region typically corresponds to the index range of the ideal parameters in the traditional ant colony algorithm, indicating that the dynamic optimization of the global path search in ant colonies is constrained by the trade-off between its prior and stochastic parameters.

[0082] To achieve feature extraction of ideal hyperparameters, a corresponding second-order difference curve is generated according to Formula 8 to explicitly represent the change vector of the first-order change in the path search. For example... Figure 5 As shown, the overall trend of the second-order difference curve, represented by the solid line, is basically consistent with that of the first-order difference curve, and it is easier to identify features—the second-order difference curve allows for more accurate segmentation of fluctuation amplitudes and enables the location and extraction of characteristic extreme points of the first-order change curve. In summary... Figure 4 and Figure 5 The path search random characteristic curve is used as a joint constraint condition with the path search distance and its corresponding change. In this example, based on the determination of the effective parameter feature point selection range, near the second wave peak where the fluctuation tends to be gentle, the third feature point corresponds to the stable amplitude segment of the second difference curve. The ant colony path search effect is optimal under the hyperparameter mapping corresponding to this point.

[0083] Based on the above conclusions regarding ideal hyperparameter localization, and following the methodological framework of Equations 9-14, the hyperparameter extraction results are as follows: Figure 6 The diagram illustrates the ideal hyperparameter indexing process in sequence: Support Vector Machine (SVM) classification prediction - feedforward passband diagnostics - feature point counter indexing. The default settings for the SVM kernel function prediction parameter group are: =1, =1, =3, set the feedforward diagnostic passband threshold parameter group as follows: =1200, k=1, the prediction and classification error is further improved. By implementing the feature point counting index method on the stable fluctuation state, the adaptive optimization solution of the ant colony β factor can be effectively derived.

[0084] Corresponding to the above optimization results, Figure 7 This paper presents a dynamic adaptive optimization curve based on the native heterogeneous competition-cooperation mechanism of ant colonies. The curve is divided into three stages: native heterogeneous regulation, hyperparameter competition extraction, and ant colony cooperative pathfinding. It is evident that the pheromone interaction and iterative update mechanism among ants is activated immediately after the optimal hyperparameters are determined, and the mapping process of the ideal hyperparameters effectively suppresses local optima. Specifically, the program sets the sum of the number of observations for native heterogeneous hyperparameter regulation and the number of cooperative iterations to 300, ensuring coverage of the effective iteration cycle of the cooperative stage. In this example scenario, the default settings for each hyperparameter are α=1, ρ=0.3, and num=5. The optimized ideal mapping β factor is β=3. After effective iteration by the ant colony, the ideal global path solution is obtained. At this point, the effective iteration count is 100, and the path distance is 7601. The global planning is illustrated below. Figure 8 As shown, the dashed line represents the global motion path. The convergence path cost and convergence effect obtained demonstrate the accuracy and effectiveness of the method described in this invention.

[0085] Combination Figure 9 As shown, a laser cleaning path planning system of the present invention includes: The imaging acquisition and processing unit is used to acquire images of the surface of the workpiece to be cleaned, perform coordinate registration and binarization processing, and obtain a binarized image of the surface of the workpiece to be cleaned. The local path determination unit is used to regularly partition the surface of the workpiece to be cleaned according to the binarized image to obtain multiple sub-partitions, and to perform path coverage for each sub-partition to obtain the local path of each sub-partition. A modeling unit is used to map the starting point of the local path in each sub-partition to obtain a global representative point set of all local paths, and to generate a traveling salesman problem model based on the global representative point set. The solving unit is used to solve the traveling salesman problem model using the native heterogeneous competition-cooperative ant colony optimization algorithm to obtain the optimal global access sequence for all the local paths; The path generation unit is used to generate a laser cleaning scanning path for the surface of the workpiece to be cleaned based on the optimal global access sequence.

[0086] The laser cleaning path planning system of the present invention has the same advantages over the prior art as the laser cleaning path planning method described above, and will not be repeated here.

[0087] In combination Figure 10 The electronic device of the present application comprises a memory and a processor; The memory is configured to store a computer program; The processor is configured to implement the laser cleaning path planning method as described above when executing the computer program.

[0088] The advantages of the electronic device of the present application are the same as those of the laser cleaning path planning method of the present application, which will not be repeated here.

[0089] Although the present application is disclosed as above, the protection scope of the present application is not limited to this. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application, and these changes and modifications shall fall within the protection scope of the present application.

Claims

1. A method of laser cleaning path planning, characterized in that, The method comprises the following steps: image acquisition and coordinate registration and binarization processing are performed on the surface of the workpiece to be cleaned to obtain a binarized image of the surface of the workpiece to be cleaned; the surface of the workpiece to be cleaned is regularly partitioned according to the binarized image to obtain a plurality of sub-partitions, and path coverage is performed for each sub-partition to obtain a local path of each sub-partition; the starting points of the local paths of each sub-partition are mapped to obtain a global representative point set of all the local paths, and a traveling salesman problem model is generated according to the global representative point set; the traveling salesman problem model is solved by a native heterogeneous competition-cooperation ant colony optimization algorithm to obtain an optimal global access sequence of all the local paths; a laser cleaning scanning path of the surface of the workpiece to be cleaned is generated according to the optimal global access sequence.

2. The laser cleaning path planning method of claim 1, wherein, The image acquisition and coordinate registration and binarization processing performed on the surface of the workpiece to be cleaned to obtain a binarized image of the surface of the workpiece to be cleaned comprises the following steps: digital imaging is performed on the surface of the workpiece to be cleaned to obtain an original gray-scale image of the surface of the workpiece to be cleaned; coordinate conversion algorithm is used to register the original gray-scale image with the actual spatial position of the surface of the workpiece to be cleaned to obtain a registered gray-scale image; adaptive threshold segmentation or maximum inter-class variance segmentation is performed on the registered gray-scale image to obtain the binarized image of the surface of the workpiece to be cleaned.

3. The laser cleaning path planning method of claim 1, wherein, The regular partitioning of the surface of the workpiece to be cleaned according to the binarized image to obtain a plurality of sub-partitions comprises the following steps: connected component analysis is performed on the foreground pixels in the binarized image to obtain a plurality of cleaning connected components; for each cleaning connected component, the minimum circumscribed contour region of the cleaning connected component is extracted; a partition size threshold is determined according to the field lens scanning range of the laser cleaning device, the cleaning accuracy requirement and the path planning efficiency requirement; the cleaning connected components are adaptively divided according to the partition size threshold to obtain a plurality of sub-partitions.

4. The laser cleaning path planning method of claim 3, wherein, The path coverage performed for each sub-partition to obtain a local path of each sub-partition comprises the following steps: the scanning reference direction of the local path is determined according to the shape of the sub-partition and the scanning direction adaptability of the laser cleaning device; the starting point, the ending point and the path node sequence in the sub-partition are obtained according to the scanning reference direction and the path coverage algorithm corresponding to the shape of the sub-partition; an initial local path of each sub-partition is generated according to the starting point, the ending point and the path node sequence in the sub-partition; the initial local path is transitionally optimized by a Bezier curve to obtain the local path.

5. The laser cleaning path planning method of claim 4, wherein, The mapping of the starting points of the local paths of each sub-partition to obtain a global representative point set of all the local paths comprises the following steps: the local coordinates of the starting points of each sub-partition in the local coordinate system of the sub-partition are obtained; the coordinate registration relationship between the image pixels and the actual spatial position of the sub-partition is determined according to the local coordinates of the starting points. According to the coordinate registration relationship, the local coordinates of the starting points are converted into global coordinates in a global coordinate system; The global coordinates of the starting points of all sub-partitions are summarized to form the global representative point set.

6. The laser cleaning path planning method of claim 1, wherein, The generation of the traveling salesman problem model according to the global representative point set comprises: Each starting point in the global representative point set is taken as a node in the traveling salesman problem model, and the Euclidean distance between any two nodes is calculated; The Euclidean distance is taken as the path cost between the two nodes, all the nodes are traversed, and the total path cost is determined; The traveling salesman problem model is constructed with the optimization goal of minimizing the total path cost, and the traveling salesman problem model comprises a node set, a path cost matrix and a traversal constraint condition.

7. The laser cleaning path planning method of claim 6, wherein, The traveling salesman problem model is solved by the native isomeric competition-cooperation ant colony optimization algorithm to obtain an optimal global access sequence of all the local paths, which comprises: A parameter mapping table of an expected heuristic factor is constructed according to the ant colony population size, the initial pheromone concentration, the pheromone evaporation factor and the preset regulation threshold and preset grouping step coefficient of the native isomeric competition stage; In the native isomeric competition, the initial pheromone concentration is kept unchanged, the expected heuristic factor is sequentially injected into the ant colony population in the order of the parameter mapping table to drive the ants to independently complete path construction, and the total path cost, the first-order difference and the second-order difference of the total path cost are obtained; When the iteration number of the ant colony population reaches the preset regulation threshold, the second-order difference is classified by support vector machine to obtain a fluctuation convergence segment, and the expected heuristic factor corresponding to the minimum first-order difference in the fluctuation convergence segment is taken as an optimal expected heuristic factor; The optimal global access sequence is obtained by ant colony iteration according to the optimal expected heuristic factor and in combination with a probability transfer algorithm.

8. The laser cleaning path planning method of claim 6, wherein, The generation of the laser cleaning scanning path of the surface of the workpiece to be cleaned according to the optimal global access sequence comprises: The nodes in the optimal global access sequence are sequentially mapped into the execution order of the nodes corresponding to the sub-partitions; The local paths of each sub-partition are sequentially extracted in the execution order of the sub-partitions, and the local paths are spliced to obtain a spliced overall path; The laser cleaning scanning path is generated by speed planning of the spliced overall path.

9. A laser cleaning path planning system, characterized by, It comprises: An imaging acquisition and processing unit configured to image and acquire the surface of the workpiece to be cleaned, perform coordinate registration and binary processing, and obtain a binary image of the surface of the workpiece to be cleaned; A local path determination unit configured to divide the surface of the workpiece to be cleaned into a plurality of sub-partitions according to the binary image, and determine a local path for each sub-partition; A modeling unit configured to map starting points of the local paths of each sub-partition to obtain a global representative point set of all the local paths, and generate a traveling salesman problem model according to the global representative point set. A solving unit is configured to solve the TSP model by using a native heterogeneous competition-cooperation ant colony optimization algorithm to obtain an optimal global visiting sequence of all the local paths. A path generating unit is configured to generate a laser cleaning scanning path for the surface of the workpiece to be cleaned according to the optimal global visiting sequence.

10. An electronic device, comprising: comprising a memory and a processor; The memory is configured to store a computer program; The processor is configured to implement the laser cleaning path planning method according to any one of claims 1 to 8 when executing the computer program.

Images