Method and system for generating nozzle operating trajectory based on mechanical arm driving

Abstract

The application discloses a nozzle operation track generation method and system based on mechanical arm driving, relates to the field of track generation methods, and comprises the following steps: scanning a part, obtaining picture information of the part, obtaining edge data of a work surface based on the picture information, obtaining shape data of the work surface based on the edge data, obtaining a centroid of the work surface based on the shape data, and establishing a plane coordinate system based on the centroid, performing grid division on the work surface based on the shape data and the plane coordinate system, judging whether there is an isolated point, obtaining coordinate information of the isolated point if the isolated point exists, taking the isolated point as a starting point to generate an initial motion track, performing track trimming based on the initial motion track to obtain a motion track, dataizing the motion track based on the motion track to obtain track data, converting the track data, generating mechanical arm action data, and outputting the mechanical arm action data. The application has the effect of improving the fluency and flexibility of the motion track of the nozzle.

Description

Technical Field

[0001] This application relates to the field of trajectory generation methods, and in particular to a method and system for generating nozzle movement trajectories based on robotic arm drive. Background Technology

[0002] In the production and manufacturing of automobile, ship, and aircraft parts, it is necessary to treat the surface of the parts. This usually involves material spraying or water jet surface strengthening. By spraying materials on the surface of the parts, it is possible to effectively prevent rust and corrosion. Water jet surface strengthening can improve the surface condition of the parts, thereby increasing their fatigue life.

[0003] Traditional surface finishing of parts requires manual labor in harsh, unpleasant, and even toxic environments, often resulting in uneven processing and substandard quality. While robots offer more flexibility in surface finishing, such as material spraying or waterjeting, they operate according to pre-defined strategies. However, changing the part being processed necessitates modifying the strategy and adjusting equipment parameters, a cumbersome process. Furthermore, robots lack the ability to automatically recognize processing results, resulting in inflexibility with complex parts, leading to repetitive or incomplete processing and ultimately unsatisfactory outcomes.

[0004] To address these issues, we propose a method and system for generating nozzle trajectory based on robotic arm drive. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for generating nozzle trajectory based on robotic arm drive, so as to solve the problems mentioned in the background art.

[0006] Firstly, the nozzle trajectory generation method based on robotic arm drive provided in this application adopts the following technical solution:

[0007] The part is scanned to obtain the image information of the part. Based on the image information, the edge of the working surface of the part that needs to be processed is extracted to obtain the edge data of the working surface.

[0008] Based on the edge data, obtain the shape data of the working surface, obtain the centroid of the working surface based on the shape data, and establish a planar coordinate system based on the centroid;

[0009] Based on the shape data and the planar coordinate system, the working surface is divided into grids, and the image in each grid is identified to determine whether there are isolated points. If there are, the coordinate information of the isolated points is obtained.

[0010] Starting from the isolated point and based on the divided grid, an initial motion trajectory is generated, and based on the initial motion trajectory, the trajectory is adjusted to obtain the motion trajectory.

[0011] Based on the motion trajectory, the motion trajectory is digitized to obtain trajectory data, the trajectory data is converted to generate robotic arm motion data and output.

[0012] By employing the above technical solution, the parts to be processed are scanned to acquire their image information. This image information is then processed to identify isolated points with distinct features on the working surface as starting points. Based on the image information of the parts, the centroid position of the working surface is determined, a coordinate system is established, and the working surface is meshed. A motion trajectory is then generated based on the starting point and the mesh, and the trajectory is refined to ensure smoothness. Finally, the motion trajectory data is digitized and converted into motion data for the robotic arm for output. This improves the smoothness and flexibility of nozzle motion trajectory generation during parts processing.

[0013] Preferably, before the step of scanning the part to obtain image information of the part, and based on the image information, extracting the edges of the working surface of the part that needs to be processed, and obtaining the edge data of the working surface, the method further includes:

[0014] Obtain the horizontal parameters of the robotic arm, and based on the horizontal parameters, obtain the initial posture information of the robotic arm;

[0015] Determine whether the initial posture information meets the standard requirements. If it does not, then determine the posture difference between the initial posture information and the standard requirements based on the initial posture information.

[0016] Based on the posture difference, posture adjustment data is generated, and the initial posture of the robotic arm is adjusted.

[0017] By adopting the above technical solution, the horizontal parameters of the robotic arm are obtained. Based on these parameters, the initial posture of the robotic arm is determined. The initial posture is then used to determine whether the robotic arm meets the standard requirements of the operation steps. If it does not meet the requirements, the posture difference between the current posture and the standard posture is obtained. This posture difference is then digitized to obtain posture adjustment data. Finally, the initial posture of the robotic arm is adjusted based on this data. This improves the stability of the robotic arm's initial state.

[0018] Preferably, the step of scanning the part to obtain image information of the part, and based on the image information, extracting the edges of the working surface of the part that needs to be processed to obtain edge data of the working surface includes:

[0019] The part is scanned to obtain the image information of the part, and the image information is converted into a grayscale image;

[0020] Based on the grayscale image, the grayscale image is smoothed and noise in the grayscale image is removed;

[0021] Set a gradient threshold, obtain the gradient value of each pixel in the grayscale image, convert the grayscale image into a binary image, and determine whether a pixel is an edge pixel based on the gradient value;

[0022] Based on the binary image, the edge pixels are connected to obtain the edge image of the working surface of the part;

[0023] Determine whether the edge image is continuous. If it is not continuous, obtain the break position of the edge image and reconnect the break position based on the existing edge image to obtain edge data.

[0024] By employing the above technical solution, the image information of the working surface of the part is acquired, converted into a grayscale image, and noise in the grayscale image is removed. Based on the gradient values ​​of the pixels in the grayscale image, the grayscale image is converted into a binary image, and it is determined whether a pixel is an edge pixel. The identified edge pixels are connected to obtain the edge image of the working surface. The continuity of the edge image is determined, and discontinuities are filled in to obtain edge data. This improves the accuracy of edge judgment of the working surface of the part.

[0025] Preferably, the steps of obtaining shape data of the working surface based on the edge data, obtaining the centroid of the working surface based on the shape data, and establishing a planar coordinate system based on the centroid include:

[0026] Based on the edge data, obtain the shape data of the working surface, and based on the shape data, obtain the area data of the working surface;

[0027] Based on the shape data and the area data, the working surface is divided into two equal-area sections to obtain four small working surfaces with the same area.

[0028] Based on the four small working surfaces, obtain the pixel positions of the four small working surfaces, and calculate the four centroids of the four small working surfaces;

[0029] Based on the four centroids, the average value of the four centroids is calculated to obtain the centroid position of the working surface, and a planar coordinate system is established based on the centroids.

[0030] By adopting the above technical solution, the shape of the working surface is determined based on the acquired edge data, and the area of ​​the working surface is calculated. The working area is divided into four equal working surfaces. Based on the pixel position of each small working surface, the centroid of the four small working surfaces is obtained. Then, based on the positional relationship of the four centroids, the centroid position of the working surface is calculated. This improves the accuracy of determining the centroid position of the working surface.

[0031] Preferably, the step of dividing the working surface into a grid based on the shape data and the planar coordinate system, identifying the image in each grid, determining whether there are isolated points, and if so, obtaining the coordinate information of the isolated points, includes:

[0032] Based on the shape data and the planar coordinate system, the working surface is divided into a mesh to obtain the mesh of the working surface;

[0033] The image content in the grid is identified, and the edge grids containing the edge data are classified to establish an edge grid set;

[0034] Image analysis is performed on the edge grids in the edge grid set to obtain the image features of the image in the edge grids. The visibility and specificity of the image features are determined and obtained. The visibility and specificity are numerically processed to obtain the visibility value and the specificity value.

[0035] The visible value and the special value are added together to obtain the feature value of the image feature, and based on the feature value, the edge grid with the largest feature value is taken as the starting grid.

[0036] Based on the starting grid, edge features in the starting grid are obtained, and based on the edge features, isolated points are determined and their coordinate information is obtained.

[0037] By adopting the above technical solution, the working surface is divided into grids according to its shape and planar coordinate system. The content within the grids is identified, and it is determined whether it is an edge grid and categorized accordingly. Feature extraction is performed on the images within the edge grids, and the visibility and specificity of the images in each grid are assessed. The edge grid with the strongest visibility and specificity is selected as the starting grid, and the edge features of the images in the starting grid are determined to identify isolated points. This improves the globality of finding and identifying isolated points.

[0038] Preferably, the step of generating an initial motion trajectory based on the isolated point and the divided grid, and then refining the trajectory based on the initial motion trajectory to obtain the motion trajectory includes:

[0039] Based on the divided grid, trajectory points are determined in the grid, the isolated points are used as the starting point, the trajectory points are connected to obtain a trajectory straight line, and an initial motion trajectory is generated.

[0040] Obtain the processing width of the nozzle, and based on the processing width, determine whether there are processing overlaps and processing gaps in the initial motion trajectory. Mark the processing overlap area as a red area and the processing gap area as a blue area.

[0041] The red and blue areas are marked on the working surface, and the initial motion trajectory is adjusted to eliminate the red and blue areas, thereby obtaining the motion trajectory.

[0042] By adopting the above technical solution, based on the divided grid, a trajectory point is determined in each grid. Starting from an isolated point, each trajectory point is connected in series to generate an initial motion trajectory. Then, the processing width of the nozzle is obtained and mapped onto the initial motion trajectory. It is then determined whether there are areas of repeated processing or unprocessed areas on the working surface. Based on the location of these areas, the initial motion trajectory is adjusted to eliminate the areas and generate the final motion trajectory. This enhances the creativity in the motion trajectory generation process.

[0043] Preferably, the step of marking the red and blue areas on the working surface, trimming the initial motion trajectory, eliminating the red and blue areas, and obtaining the motion trajectory includes:

[0044] Obtain the processing positions of the red area and the blue area, and determine the distance between the trajectory line and the red area and the blue area;

[0045] Based on the distance, the trajectory straight line is curved by covering the blue area and removing the red area to generate a trajectory curve;

[0046] The trajectory curve is subjected to direction normalization processing to obtain the abnormal region that appears after direction normalization processing. Based on the abnormal region, the position of the trajectory points on the initial motion trajectory is adjusted to obtain the motion trajectory.

[0047] By employing the above technical solution, the location information of repeatedly processed and unprocessed areas is obtained, and the minimum distance between the trajectory line in the initial motion trajectory and each area is determined. Based on this distance, the trajectory line is curved, and the positions of trajectory points are adjusted, including position changes and trajectory point fusion, to obtain the final motion trajectory. This improves the adaptability of the motion trajectory to different problem areas during the adjustment process.

[0048] Preferably, after the steps of generating an initial motion trajectory based on the isolated point and the divided grid, and then refining the trajectory based on the initial motion trajectory to obtain the motion trajectory, the method further includes:

[0049] Based on the image information of the part, determine whether there are any different points of concavity or protrusion in the image information;

[0050] Based on the planar coordinate system, determine the position information of the different points and mark them as operation points on the motion trajectory;

[0051] Obtain the shape features of the different points, determine the processing operations for the different points based on the shape features, and add the processing operations to the marked points.

[0052] By adopting the above technical solution, based on the image information of the work surface of the part, it is determined whether there are protruding or recessed parts in the image. Based on the shape characteristics of these parts, it is determined whether to perform machining operations on these parts. At the same time, these parts are marked as operation points on the motion trajectory, and the machining operations are added to the marked points. This improves the predictability of machining special parts of the part during the machining process.

[0053] Secondly, this application provides a nozzle trajectory generation system based on a robotic arm drive, comprising:

[0054] Attitude information acquisition module: used to acquire the attitude information of the robotic arm;

[0055] Image information acquisition module: used to acquire image information of the working surface of the part;

[0056] Nozzle data acquisition module: used to acquire the processing width data of the nozzle during operation;

[0057] Data processing module: configured to be data connected to the attitude information acquisition module, the image information acquisition module, and the nozzle data acquisition module, and used to process the acquired data;

[0058] Motion trajectory generation module: configured to be data connected to the data processing module, used to generate a motion trajectory based on the data output by the data processing module;

[0059] Execution module: configured to connect to the data processing module and the motion trajectory generation module for performing operations on the posture of the robotic arm and the motion trajectory of the nozzle.

[0060] Preferably, the motion trajectory generation module includes a trajectory generation unit and a trajectory trimming unit;

[0061] The trajectory generation unit generates an initial motion trajectory based on the data output by the data processing module;

[0062] The trajectory correction unit judges the initial motion trajectory and corrects the initial motion trajectory according to the problems found, thereby obtaining the motion trajectory.

[0063] In summary, this application includes at least one of the following beneficial technical effects:

[0064] 1. The system scans the parts to be processed, acquires their image information, processes this information, identifies isolated points with distinct features on the work surface as starting points, determines the centroid of the work surface based on the part's image information, establishes a coordinate system, and meshes the work surface. A motion trajectory is then generated based on the starting point and the mesh, refined to ensure smoothness, and finally digitized and converted into motion data for the robotic arm for output. This improves the smoothness and flexibility of nozzle motion trajectory generation during part processing.

[0065] 2. Based on the shape and planar coordinate system of the working surface, the working surface is divided into grids. The content within the grids is identified, and it is determined whether it is an edge grid and categorized accordingly. Feature extraction is performed on the images within the edge grids, and the visibility and specificity of the images in each grid are assessed. The edge grid with the strongest visibility and specificity is selected as the starting grid. The edge features of the images in the starting grid are then assessed, and isolated points are identified. This improves the globality of finding and identifying isolated points.

[0066] 3. Based on the image information of the work surface of the part, determine whether there are any protruding or recessed parts on the part. Based on the shape characteristics of these parts, determine which parts will be machined, and mark these parts as operation points on the motion trajectory, adding the machining operation to the marked points. This improves the predictability of machining special parts of the part during the machining process. Attached Figure Description

[0067] Figure 1 This is a flowchart of the steps of the nozzle running trajectory generation method based on robotic arm drive of the present invention;

[0068] Figure 2This is a flowchart of the sub-steps before and after S100 in the nozzle trajectory generation method based on robotic arm drive of the present invention.

[0069] Figure 3 This is a flowchart of the sub-step S200 in the nozzle trajectory generation method based on robotic arm drive of the present invention;

[0070] Figure 4 This is a flowchart of the sub-steps S300 and S400 in the nozzle trajectory generation method based on robotic arm drive of the present invention.

[0071] Figure 5 This is a flowchart of the sub-steps after S403 and S400 in the nozzle trajectory generation method based on robotic arm drive of the present invention.

[0072] Figure 6 This is a block diagram of the nozzle trajectory generation system based on robotic arm drive according to the present invention.

[0073] Explanation of reference numerals in the attached diagram: 1. Attitude information acquisition module; 2. Image information acquisition module; 3. Nozzle data acquisition module; 4. Data processing module; 5. Motion trajectory generation module; 6. Execution module. Detailed Implementation

[0074] The following is in conjunction with the appendix Figure 1-6 This application will be described in further detail, but the embodiments of the present invention are not limited thereto.

[0075] This application discloses a method and system for generating nozzle running trajectory based on robotic arm drive.

[0076] In this embodiment, refer to Figure 1 A method for generating nozzle trajectory based on robotic arm drive includes the following steps:

[0077] S100: Scan the part to obtain the image information of the part, and based on the image information, extract the edges of the working surface of the part that needs to be processed to obtain the edge data of the working surface.

[0078] S200: Based on edge data, obtain the shape data of the working surface, based on the shape data, obtain the centroid of the working surface, and establish a planar coordinate system based on the centroid;

[0079] S300: Based on shape data and a planar coordinate system, the working surface is divided into grids, and the image in each grid is identified to determine whether there are isolated points. If so, the coordinate information of the isolated points is obtained.

[0080] S400: Starting from an isolated point and based on the divided grid, an initial motion trajectory is generated, and based on the initial motion trajectory, the trajectory is trimmed to obtain the motion trajectory;

[0081] S500: Based on the motion trajectory, the motion trajectory is digitized, the trajectory data is acquired, the trajectory data is converted, and the robotic arm motion data is generated and output.

[0082] It should be noted that the above steps are only the preferred implementation order. In the actual implementation process, some steps can be changed without affecting the overall implementation effect.

[0083] Reference Figure 2 Before step S100, the part is scanned to obtain its image information. Based on the image information, the edges of the work surface to be processed are extracted. Before the step of obtaining the edge data of the work surface, the following steps are also included:

[0084] S111: Obtain the horizontal parameters of the robotic arm, and based on the horizontal parameters, obtain the initial posture information of the robotic arm;

[0085] S112: Determine whether the initial attitude information meets the standard requirements. If not, determine the attitude difference between the initial attitude information and the standard requirements based on the initial attitude information.

[0086] S113: Based on the attitude difference, generate attitude adjustment data and adjust the initial attitude of the robotic arm.

[0087] In application, the horizontal parameters of the robotic arm are acquired, and the initial posture of the robotic arm is determined based on these parameters. The posture of the robotic arm is then adjusted according to this posture data. For example, the standard posture of a robotic arm is horizontal with the nozzle pointing vertically downwards. However, during the inspection process, it is found that the robotic arm is tilted, and the nozzle is also pointing diagonally downwards. The initial horizontal parameters of the robotic arm are then acquired, and it is determined that there is a 15-degree deviation between the initial posture and the standard posture. Based on this 15-degree deviation, the posture of the robotic arm is adjusted to restore it to the standard horizontal posture.

[0088] Reference Figure 2 In step S100, the part is scanned to obtain its image information. Based on the image information, the edges of the work surface to be processed are extracted to obtain the edge data of the work surface. This step includes:

[0089] S101: Scan the part to obtain its image information and convert it into a grayscale image;

[0090] S102: Based on the grayscale image, smooth the grayscale image and remove noise from the grayscale image;

[0091] S103: Set the gradient threshold, obtain the gradient value of each pixel in the grayscale image, convert the grayscale image into a binary image, and determine whether a pixel is an edge pixel based on the gradient value.

[0092] S104: Based on the binary image, edge pixels are connected to obtain the edge image of the working surface of the part;

[0093] S105: Determine whether the edge image is continuous. If it is not continuous, obtain the break position of the edge image and reconnect the break position based on the existing edge image to obtain edge data.

[0094] In application, the part is scanned to obtain its image information, which is then processed to obtain the edge data. For example, if a part is a cube, and the image information of one side of the part is obtained, the image information is converted into a grayscale image and denoised. At the same time, the gradient value of each pixel in the grayscale image is obtained, and the grayscale image is converted into a binary image. Based on the magnitude of the gradient value, it is determined which pixel belongs to the edge pixel, the edge pixel is drawn, and if it is determined that a certain edge is not continuous, the edge is continued to obtain the edge data of the cube part image.

[0095] Reference Figure 3 In step S200, the steps of obtaining the shape data of the working surface based on the edge data, obtaining the centroid of the working surface based on the shape data, and establishing a planar coordinate system based on the centroid include:

[0096] S201: Based on edge data, obtain the shape data of the working surface, and based on the shape data, obtain the area data of the working surface;

[0097] S202: Based on shape data and area data, the working surface is divided into two equal-area sections to obtain four small working surfaces with the same area;

[0098] S203: Based on the four small working surfaces, obtain the pixel positions of the four small working surfaces and calculate the four centroids of the four small working surfaces;

[0099] S204: Based on the four centroids, calculate the average value of the four centroids, obtain the centroid position of the working surface, and establish a planar coordinate system based on the centroid.

[0100] In application, the shape of the working surface is determined based on edge data, and it is divided into four equal parts. The centroids of the four parts are obtained, and the centroid position is calculated based on the centroids. For example, if the working surface of a part is an irregular shape, its shape is determined based on the edge data of the part, and it is then divided into four smaller working surfaces of equal area based on its size. The pixels of the smaller working surfaces are processed, the centroid positions of the four smaller working surfaces are calculated, and the centroid position of the irregular working surface is calculated based on the positions of the four centroids. At the same time, a planar coordinate system is established based on the centroid position.

[0101] Reference Figure 4 In step S300, the working surface is divided into grids based on shape data and a planar coordinate system, and the image in each grid is identified to determine whether there are isolated points. If so, the coordinate information of the isolated points is obtained. This step includes:

[0102] S301: Based on shape data and a planar coordinate system, the working surface is meshed to obtain the mesh of the working surface;

[0103] S302: Identify the image content in the grid, classify the edge grids containing edge data, and establish an edge grid set;

[0104] S303: Perform image analysis on the edge grids in the edge grid set, obtain the image features of the image in the edge grids, determine and obtain the visibility and specificity of the image features, perform numerical processing on the visibility and specificity, and obtain the visibility value and specificity value.

[0105] S304: Add the visible value and the special value to obtain the feature value of the image feature, and based on the feature value, take the edge grid with the largest feature value as the starting grid.

[0106] S305: Based on the starting grid, obtain the edge features in the starting grid, determine isolated points based on the edge features, and obtain the coordinate information of the isolated points.

[0107] In application, the working surface is divided into grids based on its shape and planar coordinate system. Edge grids containing the lake's edge are then identified within the grid, and the characteristics of the images within these edge grids are determined. A point with these distinctive features is designated as the starting point and thus an isolated point. For example, on an irregular working surface, grid division and edge grid identification are performed. The content of these edge grids is then identified. If one grid's edge image is found to be unique, containing a sharp corner, while the images in other edge grids are all line segments or curves, then the grid corresponding to this sharp corner is chosen as the starting grid. Simultaneously, the edge features of the images within this starting grid are identified, and the tip of the sharp corner is designated as the isolated point, and its coordinates are obtained.

[0108] Reference Figure 4In step S400, the steps of generating an initial motion trajectory based on an isolated point and a divided mesh, and then refining the trajectory based on the initial motion trajectory to obtain the motion trajectory include:

[0109] S401: Based on the divided grid, determine the trajectory points in the grid, take isolated points as the starting point, connect the trajectory points to obtain the trajectory line, and generate the initial motion trajectory;

[0110] S402: Obtain the processing width of the nozzle. Based on the processing width, determine whether there are processing overlaps and missing processing positions in the initial motion trajectory. Mark the processing overlap area as a red area and the processing missing area as a blue area.

[0111] S403: Mark the red and blue areas on the working surface, and perform trajectory correction on the initial motion trajectory, eliminating the red and blue areas to obtain the motion trajectory.

[0112] In application, based on the divided grid, trajectory points are determined, and an initial motion trajectory is established starting from isolated points. The processing effect is then assessed, and the initial motion trajectory is adjusted accordingly. For example, if an initial motion trajectory is established on an irregular working surface, and the processing width of the nozzle is found to be 4 cm, this 4 cm width is incorporated into the initial motion trajectory. If multiple areas of repeated processing and unprocessed areas are found, the initial motion trajectory is adjusted based on the location of these areas to obtain the processing motion trajectory for the irregular working surface.

[0113] Reference Figure 5 In step S403, the steps of marking the red and blue areas on the working surface, trimming the initial motion trajectory, eliminating the red and blue areas, and obtaining the motion trajectory include:

[0114] S4031: Obtain the processing positions of the red and blue zones, and determine the distance between the trajectory line and the red and blue zones;

[0115] S4032: Based on distance, the trajectory straight line is processed into a curve, the blue area is covered and the red area is removed to generate a trajectory curve;

[0116] S4033: Perform direction normalization processing on the trajectory curve, obtain the abnormal region that appears after direction normalization processing, and adjust the position of the trajectory points on the initial motion trajectory based on the abnormal region to obtain the motion trajectory.

[0117] In application, based on the identified unprocessed and repeatedly processed areas, and according to the positional relationship between these areas and the initial trajectory, the route and position of the trajectory points are adjusted. For example, if there is an unprocessed area and a repeatedly processed area, located on opposite sides of a trajectory line, the trajectory line is curved, moving part of the line from the repeatedly processed area to the unprocessed area. Simultaneously, if two adjacent grid points are too close together, resulting in repeated areas during processing, these two trajectory points are merged into a single trajectory point.

[0118] Reference Figure 5 After step S400, starting from the isolated point and based on the divided mesh, an initial motion trajectory is generated, and based on the initial motion trajectory, trajectory trimming is performed. After obtaining the motion trajectory, the process also includes:

[0119] S411: Based on the image information of the part, determine whether there are any differences in the image information, such as depressions or protrusions;

[0120] S412: Based on the planar coordinate system, determine the position information of different points and mark them as operation points on the motion trajectory;

[0121] S413: Obtain the shape features of different points, and based on the shape features, determine the processing operations for different points, and add the processing operations to the marked points.

[0122] In application, the image information of the part is processed to determine whether there are protrusions or depressions on the working surface of the part that require special processing operations. These locations are then marked on the motion trajectory, and the operation content is added to the marked points. In another application, if a part has multiple protrusions on its surface, and special operations are required to process these protrusions to ensure processing results, the position information of these protrusions is acquired, and marked points are added to the motion trajectory. Simultaneously, the operation content for processing these protrusions is acquired and added to the marked points. When the nozzle processes to the marked point along the motion trajectory, the operation content in the marked point is read. Based on the operation content, the robotic arm and nozzle are controlled to complete the processing operation, and then the regular processing continues along the motion trajectory.

[0123] Reference Figure 6 A nozzle trajectory generation system based on a robotic arm includes:

[0124] Attitude information acquisition module 1: Used to acquire the attitude information of the robotic arm;

[0125] Image information acquisition module 2: Used to acquire image information of the working surface of the part;

[0126] Nozzle data acquisition module 3: Used to acquire the processing width data of the nozzle during operation;

[0127] Data processing module 4: configured to connect with attitude information acquisition module 1, image information acquisition module 2, and nozzle data acquisition module 3, and to process the acquired data;

[0128] Motion trajectory generation module 5: configured to connect to the data processing module 4 for generating motion trajectories based on the data output by the data processing module 4;

[0129] Execution module 6: Configured to connect with data processing module 4 and motion trajectory generation module 5 to perform related operations on the posture of the robotic arm and the motion trajectory of the nozzle.

[0130] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for generating nozzle motion trajectory based on robotic arm drive, characterized in that, Includes the following steps: The part is scanned to obtain the image information of the part. Based on the image information, the edge of the working surface of the part that needs to be processed is extracted to obtain the edge data of the working surface. Based on the edge data, obtain the shape data of the working surface, obtain the centroid of the working surface based on the shape data, and establish a planar coordinate system based on the centroid; Based on the shape data and the planar coordinate system, the working surface is divided into grids, and the image in each grid is identified to determine whether there are isolated points. If there are, the coordinate information of the isolated points is obtained. Starting from the isolated point and based on the divided grid, an initial motion trajectory is generated, and based on the initial motion trajectory, trajectory refinement is performed to obtain the final motion trajectory, including: Based on the divided grid, trajectory points are determined in the grid, the isolated points are used as the starting point, the trajectory points are connected to obtain a trajectory straight line, and an initial motion trajectory is generated. Obtain the processing width of the nozzle, and based on the processing width, determine whether there are processing overlaps and missing processing positions in the initial motion trajectory. Mark the processing overlap area as a red area and the processing missing area as a blue area. The red and blue areas are marked on the working surface, and the initial motion trajectory is adjusted by eliminating the red and blue areas to obtain the motion trajectory, including: Obtain the processing positions of the red area and the blue area, and determine the distance between the trajectory line and the red area and the blue area; Based on the distance, the trajectory straight line is curved by covering the blue area and removing the red area to generate a trajectory curve; The trajectory curve is subjected to direction normalization processing to obtain the abnormal region that appears after direction normalization processing. Based on the abnormal region, the position of the trajectory points on the initial motion trajectory is adjusted to obtain the motion trajectory. Based on the motion trajectory, the motion trajectory is digitized to obtain trajectory data, the trajectory data is converted to generate robotic arm motion data and output.

2. The nozzle motion trajectory generation method based on robotic arm drive according to claim 1, characterized in that, Before the steps of scanning the part to obtain image information of the part, extracting the edges of the working surface of the part to be processed based on the image information, and obtaining the edge data of the working surface, the method further includes: Obtain the horizontal parameters of the robotic arm, and based on the horizontal parameters, obtain the initial posture information of the robotic arm; Determine whether the initial posture information meets the standard requirements. If it does not, then determine the posture difference between the initial posture information and the standard requirements based on the initial posture information. Based on the posture difference, posture adjustment data is generated, and the initial posture of the robotic arm is adjusted.

3. The nozzle motion trajectory generation method based on robotic arm drive according to claim 1, characterized in that, The steps of scanning the part to obtain image information of the part, and extracting the edges of the working surface of the part to be processed based on the image information to obtain the edge data of the working surface include: The part is scanned to obtain the image information of the part, and the image information is converted into a grayscale image; Based on the grayscale image, the grayscale image is smoothed and noise in the grayscale image is removed; Set a gradient threshold, obtain the gradient value of each pixel in the grayscale image, convert the grayscale image into a binary image, and determine whether a pixel is an edge pixel based on the gradient value; Based on the binary image, the edge pixels are connected to obtain the edge image of the working surface of the part; Determine whether the edge image is continuous. If it is not continuous, obtain the break position of the edge image and reconnect the break position based on the existing edge image to obtain edge data.

4. The nozzle motion trajectory generation method based on robotic arm drive according to claim 1, characterized in that, The steps of obtaining the shape data of the working surface based on the edge data, obtaining the centroid of the working surface based on the shape data, and establishing a planar coordinate system based on the centroid include: Based on the edge data, obtain the shape data of the working surface, and based on the shape data, obtain the area data of the working surface; Based on the shape data and the area data, the working surface is divided into two equal-area sections to obtain four small working surfaces with the same area. Based on the four small working surfaces, obtain the pixel positions of the four small working surfaces, and calculate the four centroids of the four small working surfaces; Based on the four centroids, the average value of the four centroids is calculated to obtain the centroid position of the working surface, and a planar coordinate system is established based on the centroids.

5. The nozzle motion trajectory generation method based on robotic arm drive according to claim 1, characterized in that, The step of dividing the working surface into a grid based on the shape data and the planar coordinate system, identifying the image in each grid, determining whether there are isolated points, and if so, obtaining the coordinate information of the isolated points, includes: Based on the shape data and the planar coordinate system, the working surface is divided into a mesh to obtain the mesh of the working surface; The image content in the grid is identified, and the edge grids containing the edge data are classified to establish an edge grid set; Image analysis is performed on the edge grids in the edge grid set to obtain the image features of the image in the edge grids. The visibility and specificity of the image features are determined and obtained. The visibility and specificity are numerically processed to obtain the visibility value and the specificity value. The visible value and the special value are added together to obtain the feature value of the image feature, and based on the feature value, the edge grid with the largest feature value is taken as the starting grid. Based on the starting grid, edge features in the starting grid are obtained, and based on the edge features, isolated points are determined and their coordinate information is obtained.

6. The nozzle motion trajectory generation method based on robotic arm drive according to claim 1, characterized in that, After the steps of generating an initial motion trajectory based on the isolated point and the divided grid, and then refining the trajectory based on the initial motion trajectory to obtain the motion trajectory, the method further includes: Based on the image information of the part, determine whether there are any different points of concavity or protrusion in the image information; Based on the planar coordinate system, determine the position information of the different points and mark them as operation points on the motion trajectory; Obtain the shape features of the different points, determine the processing operations for the different points based on the shape features, and add the processing operations to the marked points.

7. A nozzle motion trajectory generation system based on robotic arm drive, using the nozzle motion trajectory generation method based on robotic arm drive as described in any one of claims 1-6, characterized in that, include: Posture information acquisition module (1): used to acquire the posture information of the robotic arm; Image information acquisition module (2): used to acquire image information of the working surface of the part; Nozzle data acquisition module (3): used to acquire the processing width data of the nozzle during operation; Data processing module (4): configured to work with the attitude information acquisition module (1) and the image information The acquisition module (2) and the nozzle data acquisition module (3) are connected for data processing of the acquired data; Motion trajectory generation module (5): configured to be data connected to the data processing module (4) and used to generate a motion trajectory based on the data output by the data processing module (4); Execution module (6): configured to connect with the data processing module (4) and the motion trajectory generation module (5) for performing operations on the posture of the robotic arm and the motion trajectory of the nozzle.

8. The nozzle motion trajectory generation system based on robotic arm drive according to claim 7, wherein... The features include: The motion trajectory generation module (5) includes a trajectory generation unit and a trajectory trimming unit; The trajectory generation unit generates an initial motion trajectory based on the data output by the data processing module (4); The trajectory correction unit judges the initial motion trajectory and adjusts it according to the problems found. The initial motion trajectory is adjusted to obtain the final motion trajectory.

Images