A continuous painting method of an automatic painting robot

Abstract

The application relates to the field of spraying technology, in particular to a continuous spraying method of an automatic spraying robot, which comprises the following steps: S1, moving according to the set moving position of a cloud processing module and executing the set spraying gun action; S2, the cloud processing module records the moving position of the spraying robot when the spraying robot stops; S3, the spraying robot is guided to the stopped moving position; S4, an image is shot by a camera module and uploaded to the cloud processing module for edge detection; S5, a laser detector is used to verify the detected edge; S6, a spraying area image with the clearly divided sprayed area and unsprayed area is obtained according to the verified edge; S7, the position of the unsprayed area is analyzed, and a signal is sent to the spraying robot to execute the spraying operation. The application realizes continuous spraying operation and improves the efficiency by using the double verification mode of image edge detection and laser ranging to accurately judge the spraying position when the previous spraying robot stops spraying and continue spraying.

Description

Technical Field

[0001] This invention relates to the field of spraying technology, specifically to a continuous spraying method using an automated spraying robot. Background Technology

[0002] As specialized industrial robots, painting robots significantly reduce the labor intensity of workers and improve efficiency. Existing painting robots use spray guns fixedly mounted inside the factory via a movable arm. The workpiece is conveyed to the range of the movable arm, where the painting robot performs the spraying according to the pre-programmed spray gun movements.

[0003] With the development of positioning and navigation technology, freely moving painting robots have emerged. Functions such as detecting spatial range using laser detectors and sensing obstacles and adjusting direction for obstacle avoidance using cameras or other sensors have been achieved. However, current technologies only consider completing all painting operations in a single process. What happens when the painting operation is halfway through, such as insufficient paint material, insufficient power for the painting robot, the need for multiple painting robots to take turns painting, or an external force causing the painting robot to suddenly stop? Current technologies do not consider how to continue painting at the position where the previous painting robot stopped. In particular, the spray gun of the painting robot can move at multiple angles; that is, when the surface to be painted is not on the same plane as the robot's movement plane, positioning devices alone cannot achieve the effect of continuing painting. Summary of the Invention

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a continuous spraying method for an automatic spraying robot, comprising the following steps: S1, The painting robot is equipped with a positioning device and a wireless network connection device. The wireless network connection device is connected to the cloud processing module. The cloud processing module guides the painting robot to move to the set position according to the set path. The spray gun of the painting robot is connected to the cloud processing module through the wireless network connection device. The spray gun performs spraying operations in the corresponding spraying area according to the spray gun action set by the cloud processing module. The spraying area is the area to be sprayed. S2, When the painting robot stops painting, the cloud processing module records the current movement position of the painting robot; S3, the cloud processing module sends the previous painting robot's position when it stopped painting to the painting robot via wireless network signal, guides the painting robot to the designated position, and determines whether the painting robot has reached the designated position through the painting robot's positioning device; S4, after reaching the designated moving position, the camera module on the painting robot captures an image of the painting area corresponding to the moving position, and sends the image of the painting area to the cloud processing module through the wireless network connection device. After image analysis, the cloud processing module determines the edges of the painting area and generates a painting area image with the edges. S5, the laser detector on the painting robot scans both sides of the edge obtained in step S4, and confirms the authenticity of the edge by detecting the distance information between the two sides of the edge. The logical formula is as follows: ,in For the set target value, This is the distance data on one side of the edge. This is the distance data on the other side of the edge; S6, the cloud processing module will meet... By combining the edge values ​​with the image information of the sprayed area containing the edges, a sprayed area image is obtained that clearly delineates the sprayed and unsprayed areas; S7, the cloud processing module calculates the spray gun actions that the spray gun should perform based on the spray area images of the divided sprayed and unsprayed areas, and sends the corresponding spraying instructions to the spraying robot.

[0005] Preferably, the painting robot sends its location information to the cloud processing module every time it changes position.

[0006] Preferably, the camera module in step S4 is a camera, which is connected to the cloud processing module via a wireless network connection device. The cloud processing module performs image analysis as follows.

[0007] Grayscale conversion involves processing images captured by a camera to obtain a grayscale image.

[0008] Gaussian filtering is used to process grayscale images by applying Gaussian filtering to reduce noise or outliers, making the grayscale image data smoother and more continuous.

[0009] Contrast enhancement: By using histogram equalization, the contrast of grayscale images is improved, making the contrast between bright and dark areas more distinct.

[0010] Edge detection uses edge detection algorithms to detect edges in an image and extract edge information.

[0011] Edge marking marks the location of the extracted edges in the sprayed area image.

[0012] Preferably, edge detection algorithms include the Sobel algorithm, the Prewitt algorithm, and the Roberts algorithm.

[0013] Preferably, in step S5 The value is set to the coating thickness of the sprayed paint.

[0014] Preferably, in step S6, the calculated value obtained from... The defined edges are combined with the images captured by the camera module, based on... and The data is used to determine the areas that have been painted and those that have not.

[0015] Preferably, when > hour, The area where the detection point is located is the already coated area; when < hour, The area where the detection point is located is the area that has already been sprayed.

[0016] Preferably, in step S7, based on the position of the unsprayed area in the sprayed area image, the spray gun action corresponding to the position of the unsprayed area is determined, and the spray gun action continues to be executed from the previous spray gun action of the corresponding spray gun action.

[0017] Compared with existing technologies, the continuous spraying method of this automated spraying robot has the following advantages: 1. Since the area to be sprayed may have multiple color and pattern variations, such as spraying new paint on the surface of an existing pattern, the edge obtained by the edge detection algorithm captured by the camera module cannot accurately represent the edge between the sprayed and unsprayed areas. Therefore, a laser detector is used to confirm the authenticity of the edge. Second, based on the possible unevenness of the surface of the sprayed area, such as the surface of the sprayed wall protruding outward in a stepped manner from top to bottom, the laser detector cannot determine whether the protrusion is caused by the already sprayed area or by the surface shape. The camera module is used to limit the scanning area of ​​the laser detector through the edge detection algorithm, and only the detected edges are confirmed, without directly detecting the edges. Third, by detecting the edges, the spray gun can automatically determine the spraying position when the previous spraying robot interrupted its work, and automatically continue the spraying operation from the interrupted spraying position.

[0018] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0019] Figure 1 This is a schematic flowchart of the method of the present invention; Figure 2This is a schematic diagram of an embodiment to which the method of the present invention is applicable; Figure 3 This is a schematic diagram of Embodiment 2 to which the method of the present invention is applicable; In the diagram: 1. Surface to be sprayed; 2. Moving device; 3. Movable arm; 4. Spray gun; 5. Detection module. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figures 1-3 This invention provides a technical solution: a continuous spraying method for an automatic spraying robot, comprising the following steps: S1, the painting robot is equipped with a positioning device and a wireless network connection device, which connects to the cloud processing module. The cloud processing module guides the painting robot to move to a set position according to a pre-defined path. The spray gun of the painting robot connects to the cloud processing module via the wireless network connection device, and performs painting operations within the corresponding painting area according to the spray gun actions set by the cloud processing module. The area painted by a complete set of spray gun actions constitutes the painting area. That is, the cloud processing module presets two sets of settings: the moving position and the spray gun actions. The positioning device in the painting robot is only responsible for guiding the robot to a specific moving position; the spray gun action settings cause the spray gun to execute a set painting operation according to a specific action flow, and the area painted by a complete set of spray gun actions constitutes the painting area. Based on this, each time the painting robot reaches a designated moving position, the spray gun will complete the painting operation within the painting area according to a set of pre-defined spray gun actions. S2, When the painting robot stops painting, the cloud processing module records the current movement position of the painting robot. Whether the movement position here corresponds to the movement position defined in the settings is not important. The subsequent analysis process will analyze whether the current movement position is the movement position corresponding to the settings, and move to the movement position corresponding to the settings based on the analysis results. S3, the cloud processing module sends the position of the previous spraying robot when it stopped spraying to the spraying robot via wireless network signal, guides the spraying robot to the designated position, and determines whether the spraying robot has reached the designated position through the positioning device of the spraying robot. S4. After reaching the designated moving position, the camera module on the painting robot takes an image of the painting area corresponding to the moving position and sends the image of the painting area to the cloud processing module through the wireless network connection device. After the cloud processing module analyzes the image, it determines the edges in the painting area and generates a painting area image with the edges. In step S5, the laser detector on the painting robot scans both sides of the edge obtained in step S4, confirming the authenticity of the edge by detecting the distance information between the two sides. In actual painting operations, there may be painting areas with multiple mixed patterns, such as spraying new paint on an area with an existing pattern. Therefore, the edges detected by the edge detection algorithm may include edges generated by the light and dark contrast of the original pattern, and may not necessarily be edges generated after the paint is sprayed. Further confirmation of the edges is required. The logical formula for judging the authenticity of the edge in this application is as follows: ,in For the set target value, This is the distance data on one side of the edge. This is the distance data on the other side of the edge; S6, the cloud processing module will meet... By combining the edge value with the image information of the sprayed area with the edge, a sprayed area image that clearly divides the sprayed and unsprayed areas is obtained. After removing false edges by laser ranging information, the sprayed area image only has the edge that divides the sprayed and unsprayed areas. S7, the cloud processing module calculates the spray gun actions that the spray gun should perform based on the spray area images of the divided sprayed and unsprayed areas, and sends the corresponding spraying instructions to the spraying robot.

[0022] Each time the painting robot changes its position, it sends its location information to the cloud processing module to prevent the cloud processing module from being unable to determine the position of the painting robot when it suddenly stops working.

[0023] The camera module in step S4 is a camera. The camera is connected to the cloud processing module via a wireless network connection device. The steps of image analysis performed by the cloud processing module include:

[0024] Grayscale conversion involves processing images captured by a camera to obtain a grayscale image.

[0025] Gaussian filtering is used to process grayscale images by applying Gaussian filtering to reduce noise or outliers, making the grayscale image data smoother and more continuous.

[0026] Contrast enhancement: By using histogram equalization, the contrast of grayscale images is improved, making the contrast between bright and dark areas more distinct.

[0027] Edge detection uses edge detection algorithms to detect edges in an image and extract edge information.

[0028] Edge marking marks the location of the extracted edges in the sprayed area image.

[0029] Edge detection algorithms include the Sobel algorithm, the Prewitt algorithm, and the Roberts algorithm. In this application, the edges obtained by the edge detection algorithm still need to be further verified for authenticity. Therefore, the aforementioned edge detection algorithms with higher computation speed are used.

[0030] In step S5 The coating thickness is set as the absolute value of the difference between the measured distance on one side of the edge and the measured distance on the other side of the edge.

[0031] In step S6, the calculated result is... The defined edges are combined with the images captured by the camera module, based on... and The data is used to determine the painted and unpainted areas. When > hour, The area where the detection point is located is the already coated area; when < hour, The area where the detection point is located is the coated area. That is, the area with the larger distance value measured by the laser detector is the uncoated area, and the area with the smaller distance value is the coated area.

[0032] In step S7, based on the location of the unsprayed area in the sprayed area image, the spray gun action corresponding to the location of the unsprayed area is determined, and the spray gun action continues from the previous spray gun action to ensure the continuity of spraying and prevent defects such as missed spraying.

[0033] Working principle: For easier understanding, please refer to... Figure 2 .exist Figure 2 In Embodiment 1, the surface 1 to be sprayed has multiple stepped protrusions, and the edges cannot be accurately determined by the laser detector alone. The camera module and laser detector are both integrated into the detection module 5 and are aligned with the spraying direction of the spray gun 4. The spray gun 4 is connected to the moving device 2 via a movable arm 3. In this embodiment, the movable arm 3 is a telescopic cylinder, and the moving device 2 is a base connected to rollers. The base contains a positioning device, a motor that drives the rollers, and a battery that powers the motor. The cloud processing module sets the total movement distance of the spraying robot to 10 meters, advancing 1 meter at a time. The spraying action of the spray gun 4 is set to spray from top to bottom for five seconds. Therefore, the area sprayed by the spray gun 4 from top to bottom for five seconds is the spraying area. Thus, the surface 1 to be sprayed is divided into 10 spraying areas by the set movement position.

[0034] According to the above settings, the painting robot advances 1 meter at a time. After advancing 1 meter to reach the designated moving position, the spray gun 4 sprays from top to bottom for 5 seconds. After the spray gun 4 completes the 5-second spraying action, the painting robot advances another 1 meter. This cycle continues until all moving positions are completed. After each movement, the positioning device of the spray gun robot sends the position information of the moved position to the cloud processing module.

[0035] When the painting robot stops working, the cloud processing module guides the next painting robot to the designated position based on the last uploaded position information of the stopped robot. Upon reaching the designated position, the camera module on the painting robot scans the painting area corresponding to the current position, i.e., the entire range covered by the spray gun 4 during the 5-second top-to-bottom spraying in the current painting area, and generates a processed image of the painting area including the edges. The laser detector measures the distance to both sides of the edge and calculates the true edge, i.e., the edge between the painted and unpainted areas, according to the aforementioned logical formula. The cloud processing module analyzes the obtained image of the painting area with the true edges, i.e., analyzes the position of the currently unpainted area within the entire painting area, and analyzes which second of the 5-second top-to-bottom spraying range the currently unpainted area falls within, and sends a signal to the automatic painting robot to resume the painting operation from the corresponding position. For example, if the cloud processing module analyzes and determines that the currently unpainted area is located at the position that should be painted in the 3rd second of the 5-second painting range from top to bottom in the entire painting area, then it sends a signal to make the automatic painting robot continue to execute the subsequent painting actions from the painting action that should be performed in the 3rd second.

[0036] Unlike Example 1, Example 2 represents another scenario; please refer to [link / reference]. Figure 3 When the surface to be painted involves a mixture of at least two different colors, after the previous painting robot has finished painting, the next painting robot needs to continue painting based on the work already done by the previous robot. The cloud processing module controls the next painting robot to reach the position where the previous robot left off. Using the method of this invention, the next painting robot finds the end position of the previous robot's painting and continues painting from that position according to the preset spray gun actions. By reasonably setting the movement position and spray gun actions, the implementation method of Embodiment 2 can achieve the effect of automatically painting multi-colored pictures using painting robots.

[0037] Based on the above implementation scheme, the processors, modules, corresponding control programs, algorithm programs and other supporting technologies mentioned in this invention can all be implemented in conjunction with existing electrical technology, information technology, software technology and general protocols, and are not within the scope of protection claimed by this invention. This application will not describe them in detail.

[0038] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and not restrictive.

Claims

1. A continuous spraying method using an automated spraying robot, characterized in that, Includes the following steps: S1, The painting robot is equipped with a positioning device and a wireless network connection device. The wireless network connection device is connected to the cloud processing module. The cloud processing module guides the painting robot to move according to the set path and the set movement position. The spray gun of the painting robot is connected to the cloud processing module through the wireless network connection device. The spray gun performs spraying operations in the corresponding spraying range according to the spray gun action set by the cloud processing module. The spraying range of the spray gun action is the spraying area. S2, When the painting robot stops painting, the cloud processing module records the current movement position of the painting robot; S3, the cloud processing module sends the position of the previous spraying robot when it stopped spraying to the spraying robot via wireless network signal, guides the spraying robot to the designated position, and determines whether the spraying robot has reached the designated position through the positioning device of the spraying robot. S4. After reaching the designated moving position, the camera module on the painting robot takes an image of the painting area corresponding to the moving position and sends the image of the painting area to the cloud processing module through the wireless network connection device. After the cloud processing module analyzes the image, it determines the edges in the painting area and generates a painting area image with the edges. In step S5, a laser detector on the spraying robot scans both sides of the edge obtained in step S4, and confirms the authenticity of the edge by detecting the distance information on both sides of the edge. The logical formula is D m = | D l -D r |, wherein D m is a set target value, D l is distance data on one side of the edge, and D r is distance data on the other side of the edge; S6, the cloud processing module will conform to D m The edge values ​​are combined with the image information of the sprayed area with the edge to obtain a sprayed area image that clearly divides the sprayed and unsprayed areas; S7, the cloud processing module calculates the spray gun action that the spray gun should perform based on the spray area image of the divided spray area and the unsprayed area, and sends the corresponding spraying instructions to the spraying robot; D in step S5 m The value is set to the coating thickness of the sprayed paint; In step S6, the calculated value obtained from D is... m The defined edges are combined with the images captured by the camera module, according to D l and D r The data is used to determine the painted and unpainted areas; When D l >D r At that time, D r The area where the detection point is located is the already coated area; when D l <D r At that time, D l The area where the detection point is located is the area that has already been sprayed.

2. The continuous spraying method of the automatic spraying robot according to claim 1, characterized in that, In step S1, the painting robot sends its position information to the cloud processing module every time it changes its position.

3. The continuous spraying method of the automatic spraying robot according to claim 1, characterized in that, The camera module in step S4 is a camera, which is connected to the cloud processing module via a wireless network connection device. The cloud processing module performs image analysis as follows: Grayscale conversion involves processing images captured by a camera to obtain grayscale images. Gaussian filtering is used to process grayscale images by applying Gaussian filtering to reduce noise or outliers in the grayscale images, making the grayscale image data smoother and more continuous. Contrast enhancement: By using histogram equalization, the contrast of grayscale images is improved, making the contrast between bright and dark areas more distinct. Edge detection uses edge detection algorithms to detect edges in an image and extract edge information. Edge marking marks the location of the extracted edges in the image of the sprayed area.

4. The continuous spraying method of the automatic spraying robot according to claim 3, characterized in that, The edge detection algorithm includes the Sobel algorithm, the Prewitt algorithm, or the Roberts algorithm.

5. The continuous spraying method of the automatic spraying robot according to claim 1, characterized in that, In step S7, based on the position of the unsprayed area in the sprayed area image, the spray gun action corresponding to the position of the unsprayed area is determined, and the spray gun action continues to be executed from the previous spray gun action of the corresponding spray gun action.

Images