A vertical mixer blade automatic cleaning method, device, equipment and medium

Abstract

The present invention relates to the technical field of mixers. The present invention discloses a method, device, equipment and medium for automatically cleaning the blades of a vertical mixer. The method comprises: obtaining actual blade position information; adjusting the three-dimensional model of the blade from a basic position to an actual position corresponding to the actual blade position information according to the actual blade position information, and online planning a cleaning trajectory according to the three-dimensional model of the blade at the actual position; and controlling the cleaning robot to clean the blade according to the cleaning trajectory. During the blade cleaning process of the present invention, there is no need for personnel to enter the mixing workshop to manually clean the materials attached to the blades, thereby improving personnel safety and the safety of equipment use.

Description

Technical Field

[0001] The present invention relates to the technical field of mixers, and in particular to a method, device, equipment and medium for automatically cleaning blades of a vertical mixer. Background Art

[0002] Vertical mixers are primarily used to evenly mix various chemical raw materials to ultimately synthesize the desired chemical products. Because the mixing of chemical raw materials involves chemical reactions and is unstable, the mixing process is inherently dangerous. Personnel are not permitted to enter the production workshop during the paddle mixing process.

[0003] Existing vertical mixers include a mixing pot and a stirring blade that rotates relative to the pot to stir the materials within. Due to process requirements, raw materials must be added midway through the mixing process, and the paddles must be cleaned of any material adhering to or accumulating on their surfaces. This currently requires personnel to manually clean the paddles from the mixing chamber, posing a safety hazard and compromising the safety of the equipment. Summary of the Invention

[0004] In view of this, the present invention provides a method, device, equipment and medium for automatically cleaning the blades of a vertical mixer.

[0005] Specifically, the following technical solutions are included:

[0006] In a first aspect, a method for automatically cleaning blades of a vertical mixer is provided, the method comprising:

[0007] Get the actual position information of the blade;

[0008] According to the actual blade posture information, the three-dimensional model of the blade is adjusted from a basic position to an actual position corresponding to the actual blade posture information, and a cleaning trajectory is planned online based on the three-dimensional model of the blade at the actual position;

[0009] The cleaning robot is controlled to clean the blades according to the cleaning trajectory.

[0010] Preferably, before obtaining the actual blade posture information, the method includes:

[0011] A coordinate system is established with the center of the blade's revolution as the origin, and a three-dimensional model of the blade is created;

[0012] The operating parameters of the three-dimensional model of the blade are set, and the three-dimensional digital-analog information of the blade at a preset revolution angle within the revolution period is obtained by simulation to obtain a three-dimensional digital-analog information data set of the blade.

[0013] Preferably, the operating parameters of the blade three-dimensional model include the solid blade revolution speed, the solid blade rotation speed, the solid blade revolution trajectory, the solid blade rotation trajectory, the hollow blade revolution speed, the hollow blade rotation speed, the hollow blade revolution trajectory, and the hollow blade rotation trajectory.

[0014] Preferably, adjusting the three-dimensional model of the blade from a basic position to an actual position corresponding to the actual position information of the blade according to the actual position information of the blade, and planning a cleaning trajectory online according to the three-dimensional model of the blade at the actual position, includes:

[0015] Matching the actual blade pose information with the blade three-dimensional digital model information data set to obtain blade three-dimensional digital model information that matches the actual blade pose information;

[0016] Based on the three-dimensional digital model of the blade that matches the actual blade position information, the optimal deflection and translation of the three-dimensional blade model from the basic position to the actual position is calculated;

[0017] adjusting the three-dimensional model of the blade from a basic position to an actual position according to the optimal deflection and translation;

[0018] The three-dimensional model of the propeller blade at the actual position is sliced, and a normal vector of each slice is generated. A cleaning trajectory is generated online according to the normal vector of each slice.

[0019] Preferably, the actual blade posture information includes a point cloud of the actual blade posture;

[0020] The adjusting, according to the actual blade posture information, the blade three-dimensional model from a basic position to an actual position corresponding to the actual blade posture information includes:

[0021] Obtaining a basic position point cloud of the blade at a basic position according to the three-dimensional model of the blade;

[0022] Performing point cloud registration on the actual blade pose point cloud and the basic blade pose point cloud to obtain a predicted deflection and translation of the blade three-dimensional model from the basic position to the actual position;

[0023] Based on the preset motion trajectory of the blade drive shaft axis, the posture of the blade three-dimensional model adjusted to the actual position is corrected according to the predicted deflection and translation to obtain the optimal deflection and translation.

[0024] Preferably, the method of correcting the posture of the three-dimensional blade model adjusted to the actual position based on the preset blade transmission shaft axis motion trajectory and the predicted deflection and translation to obtain the optimal deflection and translation includes:

[0025] Calculating the estimated axis center position coordinates of the three-dimensional model of the blade adjusted to the actual position based on the predicted deflection and translation;

[0026] When the estimated axis center position coordinates of the blade three-dimensional model adjusted to the actual position do not overlap with the axis center motion trajectory of the blade transmission shaft, determining the matching point coordinates of the estimated axis center position coordinates of the blade three-dimensional model adjusted to the actual position in the axis center motion trajectory of the blade transmission shaft;

[0027] Calculate the optimal deflection and translation amounts according to the matching point coordinates.

[0028] Preferably, when the blades are automatically cleaned, the first blade surface to be cleaned is cleaned at the first stop time, and the second blade surface to be cleaned is cleaned at the second stop time, and the first stop time and the second stop time are adjacent to each other;

[0029] The method further comprises:

[0030] determining a first blade position and a first blade surface to be cleaned corresponding to the first parking moment based on the three-dimensional model of the blade at the first parking moment;

[0031] Determining the second blade surface to be cleaned and its position information based on the first blade position and the first blade surface to be cleaned corresponding to the first parking moment, based on a preset correspondence between the surface to be cleaned and the blade position, and a preset blade cleaning process sequence;

[0032] According to the position information of the surface to be cleaned of the second blade, controlling the blade and the three-dimensional model of the blade to move to a position corresponding to the position information of the surface to be cleaned of the second blade;

[0033] Online planning of a cleaning trajectory for the second blade surface to be cleaned based on the second blade surface to be cleaned and the three-dimensional model of the blade;

[0034] The cleaning robot is controlled to clean the surface to be cleaned of the second blade according to the cleaning trajectory of the surface to be cleaned of the second blade.

[0035] In a second aspect, a vertical mixer blade automatic cleaning device is provided, comprising:

[0036] The acquisition module is used to obtain the actual position information of the blade;

[0037] a planning module, configured to adjust the three-dimensional model of the blade from a base position to an actual position corresponding to the actual position of the blade according to the actual position of the blade, and to plan a cleaning trajectory online according to the three-dimensional model of the blade at the actual position;

[0038] The cleaning module is used to control the cleaning robot to clean the blades according to the cleaning trajectory.

[0039] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for automatically cleaning the blades of a vertical mixer as described in the first aspect are implemented.

[0040] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the automatic cleaning method of the vertical mixer blades as described in the first aspect are implemented.

[0041] The beneficial effects of the technical solution provided by the present invention include at least:

[0042] The present invention uses a posture recognition device to identify the actual posture of the blade to be cleaned. Based on the blade's actual posture, the blade's three-dimensional model is adjusted from its base position to its actual position. A cleaning trajectory is then planned online based on the blade's three-dimensional model at the actual position. A cleaning robot then cleans the blade along this trajectory. This process eliminates the need for personnel to enter the mixing room to manually remove material adhering to the blade, improving both personnel and equipment safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 This is a schematic structural diagram of an automatic cleaning system for vertical mixer blades according to one embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of the control structure of an automatic cleaning system for vertical mixer blades according to one embodiment of the present invention;

[0046] Figure 3 A schematic diagram of the image monitoring structure of an automatic cleaning system for vertical mixer blades according to one embodiment of the present invention;

[0047] Figure 4 This is a schematic diagram of the scraper structure of an embodiment of the present invention;

[0048] Figure 5 for Figure 4 Left side view of the scraper structure;

[0049] Figure 6 for Figure 4 Front view of scraper structure;

[0050] Figure 7 for Figure 4 Top view of scraper structure;

[0051] Figure 8 A schematic diagram of the connection block structure of an embodiment of the present invention;

[0052] Figure 9 This is a schematic flow chart of a method for automatically cleaning blades of a vertical mixer according to one embodiment of the present invention;

[0053] Figure 10 This is a schematic diagram of the blade structure of an embodiment of the present invention;

[0054] Figure 11 A schematic diagram of the distribution of the coordinates of the blade shoulder tip and the axis of motion according to an embodiment of the present invention;

[0055] Figure 12 A schematic diagram of the blade shoulder tip trajectory according to an embodiment of the present invention;

[0056] Figure 13 This is a schematic diagram of the point cloud registration process according to an embodiment of the present invention;

[0057] Figure 14 Schematic diagram of an automatic cleaning device for vertical mixer blades according to one embodiment of the present invention;

[0058] Figure 15 is a structural diagram of a computer device in one embodiment of the present invention;

[0059] Figure 16 FIG. 2 is another structural diagram of a computer device according to an embodiment of the present invention.

[0060] The reference numerals in the figures represent respectively:

[0061] 1-Acquisition module; 2-Planning module, 3-Cleaning module; 100-Cleaning robot; 101-Base; 102-Turntable; 103-Cantilever beam; 104-Six-axis robotic arm; 105-Six-axis force sensor; 110-Scraper; 111-Mainboard; 1111-Block; 1112-Assembly hole; 112-Mainboard sleeve; 1121-Connecting sleeve; 1122-Support plate; 1123-Third side; 1124-Fourth side; 1125-Second side; 1126 -first rib; 1127-second rib; 1128-third rib; 1129-first surface; 120-connecting block; 121-card slot; 122-connecting hole; 200-posture recognition device; 201-structured light 3D scanner; 202-backend platform; 301-paddle; 302-mixing pot; 400-control device; 500-input device; 600-temperature sensor; 700-alarm; 800-image acquisition device; 900-display.

[0062] The above drawings illustrate specific embodiments of the present invention, which will be described in more detail below. These drawings and the accompanying description are not intended to limit the scope of the present invention in any way, but rather to illustrate the concept of the present invention to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0064] Before further describing the embodiments of the present invention in detail, the directional nouns involved in the embodiments of the present invention, such as "upper", "lower", "side", etc., are used to represent the position of the subject matter. Figure 1 The directions shown in the figure are for reference only and do not limit the scope of protection of the present invention.

[0065] To make the technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0066] As mentioned above, combined with Figure 1The existing vertical mixer includes a mixing pot 302 and a paddle 301. The paddle 301 can rotate relative to the mixing pot 302 to stir the materials in the mixing pot 302. Since the process requires the addition of raw materials midway during the stirring process and the need to clean the materials adhered to or accumulated on the surface of the mixer paddle 301, this process currently requires personnel to enter the mixing workshop to manually clean the materials adhered to the paddle 301, which poses a safety hazard and reduces the safety of equipment use. To address the above technical problems, the present application proposes the following embodiments.

[0067] Example 1

[0068] like Figure 1 As shown, the vertical mixer includes a lifting device connected to the mixing pot 302. The lifting device is used to drive the mixing pot 302 to move at least in the vertical direction relative to the paddle 301, so that the mixing pot 302 can reach a first position and a second position. When the mixing pot 302 is in the first position, the paddle 301 is inside the mixing pot 302 and can rotate relative to the mixing pot 302 to stir the material in the mixing pot 302. When the mixing pot 302 is in the second position, the paddle 301 is outside the mixing pot 302, and there is a set distance between the paddle 301 and the mixing pot 302 to facilitate cleaning by the cleaning robot 100.

[0069] In this embodiment, when the mixing pot 302 reaches the second position, the automatic cleaning system can be started. The cleaning system includes a cleaning robot 100 and a posture recognition device 200. The cleaning robot 100 is located on one side of the mixer, and the posture recognition device 200 is located on the other side of the mixer. The cleaning robot 100 and the posture recognition device 200 are respectively connected to the control device 400 for communication. The posture recognition device 200 identifies the actual posture information of the blade 301 and provides it to the control device 400. The control device 400 adjusts the three-dimensional model of the blade 301 from the basic position to the actual position according to the actual posture information of the blade 301, and plans the cleaning trajectory online according to the three-dimensional model of the blade 301 at the actual position, and uses the cleaning robot 100 to clean the blade 301 according to the cleaning trajectory. As a result, there is no need for personnel to enter the mixing workshop to manually clean the materials attached to the blade 301, thereby improving the safety of equipment use. After cleaning is completed, the lifting device can drive the mixing pot 302 to move relative to the paddle 301 so that the mixing pot 302 reaches the first position, and then the paddle 301 can rotate relative to the mixing pot 302 to stir the material in the mixing pot 302.

[0070] Preferably, Figure 1As shown, the cleaning robot 100 includes a base 101, a steering structure, a six-axis robotic arm 104, and a scraper 110. The base 101 is arranged on one side of the mixer, the steering structure is arranged at the end of the base 101, one end of the six-axis robotic arm 104 is connected to the steering structure, and the other end of the six-axis robotic arm 104 is connected to the scraper 110. The steering structure is used to drive the six-axis robotic arm 104 to rotate circumferentially around the vertical axis of the base 101.

[0071] Specifically, such as Figure 1 As shown, the base 101 is positioned on the left side of the mixing pot 302, and the position recognition device 200 is positioned on the right side of the mixing pot 302. The six-axis robotic arm 104 drives the scraper 110 along the cleaning trajectory to remove material from the paddle 301. The steering mechanism allows the six-axis robotic arm 104 to be controlled to approach the paddle 301 when cleaning is required and to move away from it when cleaning is complete, preventing the cleaning robot 100 from interfering with the mixing process and potentially damaging it.

[0072] Preferably, the steering structure includes a drive motor, a turntable 102, and a cantilever beam 103. One side of the turntable 102 is connected to the base 101, the other side of the turntable 102 is connected to the first end of the cantilever beam 103, and the six-axis robotic arm 104 is connected to the second end of the cantilever beam 103. The drive motor is connected to the turntable 102, with the axis of the turntable 102 coaxially arranged with the vertical axis of the base 101. The drive motor is used to drive the turntable 102 to rotate. By rotating the cantilever beam 103 relative to the base 101 with the turntable 102 as the center, the cantilever beam 103 drives the six-axis robotic arm 104 to rotate, thereby making the six-axis robotic arm 104 more flexible.

[0073] A scraper 110 of the present application is provided with multiple cleaning structures, one cleaning structure is matched with a surface to be cleaned of the blade, and the multiple cleaning structures are used to clean the multiple surfaces to be cleaned of the blade, which simplifies the structure of the scraper 110. The scraper 110 does not need to be replaced during the cleaning process, thereby improving the cleaning efficiency. At the same time, the end connection structure of the cleaning robot is simplified.

[0074] Preferably, the multiple cleaning structures include an arc-shaped concave surface structure, an arc-shaped convex surface structure, and a flat surface structure. The arc-shaped concave surface structure is used to clean the convex surface to be cleaned of the blade, the arc-shaped convex surface structure is used to clean the concave surface to be cleaned of the blade, and the flat surface structure is used to clean the flat surface to be cleaned of the blade.

[0075] Specifically, the blades include solid blades and hollow blades. The blades of the solid blades and the hollow blades are both spiral structures. The surfaces to be cleaned of the solid blades and the hollow blades include convex surfaces, concave surfaces, shoulders and bottoms. At the same time, the shoulders and bottoms of the solid blades and the hollow blades are planar structures. The curved concave structure cleans the convex surface, the curved convex structure cleans the concave surface, and the straight structure cleans the planar structure. The blades can be cleaned by a scraper 110. The scraper 110 does not need to be replaced during the cleaning process, which improves the cleaning efficiency and simplifies the terminal connection structure of the cleaning robot.

[0076] Preferably, Figure 4 As shown, the scraper 110 includes a main structure and various cleaning structures arranged on the main structure. The main structure includes a main board 111 and a main board cover 112. The main board cover 112 is a flexible component, and the main board 111 is a rigid component. The main board cover 112 is sleeved on the main board 111, and the various cleaning structures are arranged on the main board cover 112.

[0077] Specifically, the mainboard 111 is made of metal material, which is rigid and can play a good role. The mainboard cover 112 can be made of polyurethane and anti-electric material, which has a certain flexibility and can effectively remove residual materials on the surface of the blade and avoid damaging the blade.

[0078] Preferably, motherboard sleeve 112 includes a connecting sleeve 1121 and a support plate 1122. Connecting sleeve 1121 is bag-shaped and fits over motherboard 111, forming an interference fit. Support plate 1122 is positioned on one side of connecting sleeve 1121, with the plane of support plate 1122 perpendicular to the plane of motherboard 111. Various cleaning structures are provided on support plate 1122.

[0079] Specifically, such as Figure 4 、 Figure 5 、 Figure 6 and Figure 7 As shown, the connecting sleeve 1121 is bag-shaped, with an opening at the front end. The shape of the connecting sleeve 1121 is consistent with the shape of the main board 111, and the main board 111 is inserted from the front opening of the connecting sleeve 1121. A support plate 1122 is provided at the rear end of the connecting sleeve 1121, and a variety of cleaning structures are provided on the support plate 1122.

[0080] In a specific embodiment, depending on different application scenarios, the support plate 1122 can be set at the rear end of the connecting sleeve 1121; it can also be set on the left or right side of the connecting sleeve 1121, close to the rear end of the connecting sleeve 1121.

[0081] Preferably, the support plate 1122 includes a peripheral side surface, which includes a first surface, a second surface 1125, a third surface 1123, and a fourth surface 1124. The first surface is connected to the connecting sleeve 1121; the second surface 1125 is disposed opposite the first surface and has a flat surface structure; the third surface 1123 connects the first end of the first surface and the first end of the second surface 1125 and has a curved concave surface structure; the fourth surface 1124 connects the second end of the first surface and the second end of the second surface 1125 and has a curved convex surface structure.

[0082] Specifically, such as Figure 4 As shown, the peripheral side surfaces of the support plate 1122 are the front side surface, the rear side surface, the left side surface and the right side surface of the support plate 1122. The first surface is the right side surface of the support plate 1122, which is connected to the connecting sleeve 1121. The second surface 1125 is the left side surface of the support plate 1122, which is arranged opposite to the first surface. The second surface 1125 is a flat surface structure. In order to facilitate the scraping of materials on the blades, the edges of the second surface 1125 are chamfered. The flat surface structure is used to clean the materials on the plane structure on the bottom or shoulder of the blades. The third surface 1123 is an arc-shaped concave surface structure, connecting the front end of the first surface and the front end of the second surface 1125. In order to facilitate the scraping of materials on the blades, the edges of the third surface 1123 are chamfered. The arc-shaped concave surface structure is used to clean the materials on the convex surface of the blades. The fourth surface 1124 is an arc-shaped convex structure, connecting the rear end of the first surface and the rear end of the second surface 1125. In order to facilitate scraping of materials on the blades, the edges of the fourth surface 1124 are chamfered. The arc-shaped convex structure is used to clean the materials on the concave surface of the blades.

[0083] Preferably, the support plate 1122 further includes a first side surface and a second side surface, the first side surface and the second side surface being arranged parallel to each other, with the first side surface being arranged on one side of the side circumference and the second side surface being arranged on the other side of the side circumference. A first rib 1126 is provided on the first side surface, and is arranged adjacent to the third surface 1123. The length of the first rib 1126 extends perpendicular to the plane of the main plate 111, and the surface of the first rib 1126 is a flat surface structure.

[0084] Specifically, such as Figure 4 As shown in FIG. 1 , the first side surface is the upper surface of the support plate 1122, and the second side surface is the lower surface of the support plate 1122. The extension length of the first rib 1126 is consistent with the width from the left end to the right end of the first side surface.

[0085] Specifically, such as Figures 4 to 7As shown, the connecting sleeve 1121 includes a first surface 1129, which is the upper surface of the connecting sleeve 1121. The upper surface of the support plate 1122 and the first surface 1129 of the connecting sleeve 1121 are located in the same plane and are in contact with each other. The first rib 1126 protrudes from the first surface 1129 of the connecting sleeve 1121. The surface of the first rib 1126 is a flat surface structure. To facilitate scraping of material from the blade, the top surface of the first rib 1126 is chamfered, which can easily clean material from the bottom flat structure of the blade.

[0086] Preferably, Figure 4 As shown, a second rib 1127 is further provided on the first side surface; the second rib 1127 is arranged close to the fourth surface 1124, and the length extension direction of the second rib 1127 has a certain angle with the plane of the main board 111; the surface of the second rib 1127 is a flat surface structure.

[0087] Specifically, such as Figure 4 As shown, second rib 1127 protrudes from first surface 1129 of connecting sleeve 1121. The surface of second rib 1127 is a flat structure. To facilitate scraping of material from the blade, the top surface of second rib 1127 is chamfered, making it easy to clean material from the bottom flat structure of the blade. Second rib 1127 can be provided or omitted as needed.

[0088] Furthermore, since there is a certain angle between the length extension direction of the first rib 1126 and the length extension direction of the second rib 1127, the shoulder plane or the bottom plane of the blade can be cleaned in the front-to-back direction or the left-to-right direction, such as Figure 2 The orientation shown increases the dimension of cleaning and improves the efficiency of cleaning.

[0089] Preferably, Figure 5 and Figure 6 As shown, a third rib 1128 is provided on the second side surface; the third rib 1128 and the first rib 1126 are arranged parallel to each other, and the surface of the third rib 1128 is a flat surface structure.

[0090] Specifically, third rib 1128 is provided on the lower surface of support plate 1122. The length of third rib 1128 extends perpendicular to the plane of connecting sleeve 1121 and is the same as the length of first rib 1126. Third rib 1128 has a flat surface. To facilitate scraping material off the blade, the top surface of third rib 1128 is chamfered, allowing for cleaning of material on the blade shoulder or bottom surface.

[0091] In this embodiment, the cleaning robot 100 further includes a connecting block 120 , which is disposed at the end of the six-axis robotic arm of the cleaning robot.

[0092] like Figure 8 As shown, the connecting block 120 is provided with a slot 121 and a connecting hole 122 .

[0093] like Figure 4 and Figure 8 As shown, a clamping block 1111 and an assembly hole 1112 are provided at one end of the main plate 111 of the vertical mixer blade cleaning scraper 110. When the scraper 110 is connected to the connecting block 120, the clamping slot 121 and the clamping block 1111 engage, and the connecting hole 122 and the assembly hole 1112 are provided in a one-to-one correspondence. The connecting block 120 and the main plate 111 are connected by fasteners, which can be screws.

[0094] In this embodiment, Figure 2 As shown, the automatic cleaning system for the vertical mixer blades also includes a six-dimensional force sensor 105, which is connected between the second end of the six-axis robotic arm 104 and the scraper 110. The six-dimensional force sensor 105 is in communication with the control device 400. The six-dimensional force sensor 105 is used to sense the force data of the scraper 110 and provide it to the control device 400. The control device 400 is also used to control the six-axis robotic arm 104 based on the force data to drive the scraper 110 to move and clean the blade 301. Therefore, when the force value of the scraper 110 exceeds the set value, the control device 400 can automatically control the six-axis robotic arm 104 to move so that the scraper 110 is appropriately away from the blade 301, thereby reducing the force value of the scraper 110 to avoid the scraper 110 from breaking. Of course, the control device 400 can also communicate with the alarm device. When the force value of the scraper 110 exceeds the set value, the control device 400 controls the alarm device to turn on to prompt the operator to manually control the movement of the six-axis robotic arm 104 to reduce the scraping force of the scraper 110.

[0095] Preferably, the vertical mixer blade automatic cleaning system further comprises a temperature detection element which is in communication with the control device 400 and is used to detect the temperature of the surface of the blade 301 to be cleaned.

[0096] Specifically, such as Figure 2As shown, the automatic cleaning system for vertical mixer blades also includes an alarm 700. The temperature detection element is a temperature sensor 600. The temperature detection element and the alarm 700 are respectively connected to the control device 400 for communication. The temperature detection element is used to sense the temperature data of the surface to be cleaned of the blade 301 and provide it to the control device 400. The control device 400 is used to control the activation or deactivation of the alarm 700 based on the temperature data fed back by the temperature detection element. By monitoring the temperature of the surface to be cleaned and providing an alarm, safety issues caused by excessively high temperatures of the surface to be cleaned can be effectively avoided, further improving the safety of the equipment.

[0097] Preferably, Figure 3 As shown, the automatic cleaning system for vertical mixer blades also includes an image acquisition device 800 and a display 900. The image acquisition device 800 and the display 900 are communicatively connected. The image acquisition device 800 is used to acquire image data of the mixer and the cleaning robot 100 and provide it to the display 900 for display. In this embodiment, the image data acquired by the image acquisition device 800 is displayed on the display 900, allowing the operator to monitor the cleaning process of the blade 301, determine the position of the cleaning robot 100, and confirm whether the material on the blade 301 has been completely scraped off, thereby improving the reliability of the equipment.

[0098] In this embodiment, the vertical mixer blade automatic cleaning system also includes an explosion-proof cabinet, and the control device 400 is located within the explosion-proof cabinet. This further improves the safety of the equipment. The components of the control device 400 can be explosion-proof components.

[0099] Preferably, the posture recognition device 200 includes a structured light 3D scanner 201. The posture recognition device 200 also includes a backend platform 202, the structured light 3D scanner 201 and the backend platform 202 are communicatively connected, and the backend platform 202 and the control device 400 are communicatively connected. The structured light 3D scanner 201 is used to collect the three-dimensional point cloud image of the blade 301 to be cleaned and provide it to the backend platform 202, and the backend platform 202 is used to parse the three-dimensional point cloud image to identify and obtain the spatial coordinate information of the blade 301 to be cleaned and provide it to the control device 400. In this embodiment, the structured light 3D scanner 201 is used to scan the blade 301, and there is no need to stick identification points on the blade 301. The operation is simple and the spatial coordinate information obtained by analysis is accurate. In this embodiment, the structured light 3D scanner 201 uses blue natural light for photo recognition, which meets the explosion-proof safety requirements of pyrotechnic production and has higher accuracy. Not using external identification points can avoid the contamination risk caused by the identification points.

[0100] Specifically, in this embodiment, the structured light 3D scanner 201 is positioned on one side of the mixer, relatively far from the cleaning robot 100 and the paddle 301. During the cleaning process, the structured light 3D scanner 201 and the paddle 301 do not interfere with each other, nor does it affect the accessibility of the cleaning robot 100. This also prevents contamination of the structured light 3D scanner 201 during the cleaning process. Furthermore, the structured light 3D scanner 201 requires power. In this embodiment, the power supply for the structured light 3D scanner 201 is located far from the paddle, avoiding the risk of introducing new leakage current onto the paddle surface during scraping. This is because the paddle is very sensitive to static electricity and current, thus avoiding potential safety hazards. Due to the large size of the paddle, this example places the structured light 3D scanner 201 at a certain working viewing distance, enabling it to cover the main areas of the paddle surface. The resulting point cloud is complete and formed in one go, eliminating the need for stitching, resulting in efficient point cloud generation.

[0101] In this embodiment, a point cloud registration technology with a large viewing distance is used, and the point cloud of the workpiece is registered with the blade while material is attached. Firstly, given that the blade's motion is a compound motion (while simultaneously revolving, the solid blade and the hollow blade rotate clockwise and counterclockwise, respectively, resulting in a nearly identical posture in any two parking spaces), the structured light 3D scanner 201 can identify the blade's 6Dof (xyz spatial coordinates and three-axis rotation angle) posture in any parking space within a large viewing distance of at least 1.8 meters, maintaining excellent millimeter-level recognition accuracy. Secondly, during production, the blade's outer surface is often covered with material, and registration of the blade's actual pose point cloud with the blade's basic pose point cloud enhances registration accuracy.

[0102] Preferably, the control device 400 includes a simulation module, a first matching module, a first calculation module, a second matching module, and a second calculation module. The first matching module and the first calculation module are correspondingly provided, and the second matching module and the second calculation module are correspondingly provided. The control device 400 may be provided with the first matching module, the first calculation module, the second matching module, and the second calculation module simultaneously, or may be provided with the first matching module and the first calculation module separately, or may be provided with the second matching module and the second calculation module separately.

[0103] In one embodiment, the simulation module is used to simulate and obtain a three-dimensional digital-analog information data set of the blade 301 according to the three-dimensional model of the blade 301 and preset operating parameters of the blade 301 .

[0104] like Figure 2 As shown, this embodiment may further include an input device 500 , through which the three-dimensional models of the blade 301 and the cleaning robot 100 are meshed and input into the control device 400 .

[0105] Specifically, in this embodiment, the blades 301 include solid blades and hollow blades, and the movements of the solid blades and the hollow blades include rotation and revolution. The actual posture of the blades 301 includes the actual posture of the solid blades and the actual posture of the hollow blades. In this embodiment, the cleaning trajectory is planned online by using the three-dimensional digital model information of the blades 301 corresponding to the actual posture of the blades 301. There is no need to adjust the position of the blades 301 before and after cleaning, thereby improving the efficiency of cleaning. The actual posture in this application is the posture during the first shutdown for cleaning within a processing cycle. In this application, each shutdown within a processing cycle can clean at least one surface to be cleaned of the blade.

[0106] Specifically, the preset operating parameters of the blade 301 include the solid blade revolution speed, the solid blade rotation speed, the solid blade rotation trajectory, the hollow blade revolution speed, the hollow blade rotation speed, and the hollow blade rotation trajectory. Figure 10 The solid blade rotation direction, the hollow blade rotation direction and the blade revolution direction are shown. For example, in a specific embodiment, Figure 11 Shown Figure 10 The running coordinates of points C and D on the shoulder of the solid blade, the running coordinates of points C and D on the shoulder of the hollow blade, and the blade axis coordinates ( Figure 10 axis indicated by the dashed line). Figure 12 The running trajectories of point C on the shoulder of the solid blade and point C on the shoulder of the hollow blade are shown, where t0~t 10 The trajectory line of point C on the hollow blade shoulder at the corresponding time (marked in sequence) is the red * point line, and the trajectory line of point C on the solid blade shoulder is the pink * point line. Only time t0 and time t1 are shown in the figure to indicate the order of time.

[0107] Specifically, the 3D digital-analog information dataset for blade 301 includes a 3D digital-analog information dataset for a solid blade and a 3D digital-analog information dataset for a hollow blade. In this embodiment, the 3D digital-analog information dataset for blade 301 is a point cloud dataset for each rotation of blade 301 through a preset orbital angle within the orbital period. That is, a point cloud for blade 301 is obtained for each rotation through a preset orbital angle within the orbital period. The preset orbital angle can be 1°, 5°, or 0.1°. The number of point clouds for blade 301 can be set as needed to facilitate subsequent matching.

[0108] The first matching module is used to match the actual position and posture information of the blade 301 in the three-dimensional digital and analog information data set of the blade 301 to obtain the three-dimensional digital and analog information of the blade 301 that matches the actual position and posture information of the blade 301.

[0109] The first calculation module is used to calculate the optimal deflection and translation of the three-dimensional model of the blade 301 from the basic position to the actual position based on the three-dimensional digital model information of the blade 301 that matches the actual position information of the blade 301.

[0110] Specifically, the actual blade posture information includes the actual blade posture point cloud. Based on the actual blade posture point cloud of the solid blade, the corresponding solid blade point cloud is matched in the solid blade three-dimensional digital model information dataset. Based on the corresponding solid blade point cloud, the optimal deflection and translation of the solid blade three-dimensional model from the base position to the actual position are calculated, and the solid blade cleaning trajectory is planned online. Based on the actual blade posture point cloud of the hollow blade, the corresponding hollow blade point cloud is matched in the hollow blade three-dimensional digital model information dataset. Based on the corresponding hollow blade point cloud, the optimal deflection and translation of the hollow blade three-dimensional model from the base position to the actual position are calculated, and the hollow blade cleaning trajectory is planned online.

[0111] Specifically, in this embodiment, the matching process of the actual position information of the blade 301 and the three-dimensional digital model information of the blade 301 is also a process of mutual verification of the three-dimensional digital model information of the blade 301 obtained by simulation and the collected actual position information of the blade 301.

[0112] In another embodiment, the control device 400 further includes a correction module.

[0113] The second matching module is used to perform point cloud registration between the actual blade posture information and the basic blade posture information.

[0114] Specifically, the actual blade posture information includes the actual blade posture point cloud, which is obtained by the posture recognition device 200 .

[0115] The basic blade pose information includes a point cloud representing the blade's basic position. The basic blade pose point cloud represents the point cloud when the blade is in its basic position. The basic blade position is the initial position when the blade is first put into operation.

[0116] In one embodiment, a blade base pose point cloud can be labeled in the blade's three-dimensional digital model information dataset. The blade base pose is a point cloud when the blade is in a base position. The three-dimensional model of blade 301 in the base position and the spatial coordinates of the actual blade in the base position are consistent. In another embodiment, the blade base pose point cloud can be obtained by pose recognition device 200.

[0117] The actual blade pose point cloud and the basic blade pose point cloud obtained need to be pre-processed. Figure 13 .

[0118] The second matching module and the second calculation module are based on the NDT (normal distributions transform) registration method. The NDT algorithm performs registration by maximizing the score of the source point on the probability density of the normal distribution calculated after the target point is voxelized. Because it does not require explicit calculation of the nearest corresponding points, the NDT algorithm has a fast calculation speed.

[0119] Specifically, refer to Figure 13 The second matching module is used to extract the point cloud feature points in the actual position point cloud of the blade based on a preset feature point extraction model; calculate the feature histogram of the point cloud feature points based on the preset feature extraction model; and remove the abnormal points according to the feature histogram based on the preset abnormal point removal model to obtain the initial posture of the blade at the actual position.

[0120] The second calculation module is used to calculate the predicted deflection and predicted translation of the three-dimensional model of the blade when it is adjusted from the basic position to the actual position.

[0121] refer to Figure 13 The second calculation module is used to divide the actual blade pose point cloud and the basic blade pose point cloud into grids, count the points within each grid, calculate the mean and covariance within each grid based on the number of points within each grid, and construct a Gaussian distribution for the points within each grid. It is also used to calculate the initial deflection and translation values ​​(predicted attitude) between the actual blade pose point cloud and the basic blade pose point cloud based on the initial blade pose at the actual position and the Gaussian distribution of the points within each grid. Based on the initial offset and translation values, the joint probability that the actual blade pose point cloud falls within the basic blade pose point cloud is calculated. The predicted deflection R' and translation T' are calculated based on the maximum value of these joint probabilities.

[0122] The correction module is used to correct the posture of the blade three-dimensional model adjusted to the actual position based on the pre-set blade drive shaft axis motion trajectory and the predicted deflection and predicted translation.

[0123] The correction module is used to calculate the estimated axis position coordinates of the blade three-dimensional model adjusted to the actual position based on the predicted deflection amount R' and translation amount T'. When the estimated axis position coordinates of the blade three-dimensional model adjusted to the actual position do not overlap with the axis motion trajectory of the blade transmission shaft, the matching point coordinates of the estimated axis position coordinates of the blade three-dimensional model adjusted to the actual position are determined in the axis motion trajectory of the blade transmission shaft, and the reference Figure 11 According to the coordinates of the matching points, the optimal deflection R and translation T are calculated.

[0124] The control device 400 also includes a parking planning module, in which a corresponding relationship table as shown in Table 1 is pre-set. After the cleanable surface is cleaned in the first parking, the coordinates of the subsequent parking are selected through the parking planning module to clean the remaining surface to be cleaned.

[0125] Table 1 Correspondence between parking spots and cleanable surfaces

[0126]

[0127] The control device 400 includes a planning module 2 for slicing the three-dimensional model of the blade at the actual position, generating a normal vector for each slice, and generating a cleaning trajectory online according to the normal vector for each slice.

[0128] Specifically, the planning module 2 is used to slice the surface to be cleaned, generate a normal vector for each slice, and generate a cleaning trajectory online according to the normal vector for each slice.

[0129] The control device 400 further includes a cleaning module 3 for controlling the cleaning robot to clean the blades according to the cleaning trajectory.

[0130] During automatic blade cleaning, the first blade surface to be cleaned is cleaned at the first stop, and the second blade surface to be cleaned is cleaned at the second stop. The first and second stop times are adjacent to each other. The first and second stop times correspond to two adjacent stops, not specifically the order of stops. That is, for two adjacent stops on both sides, the first stop time is the previous stop time, and the second stop time is the next stop time.

[0131] Specifically, in one embodiment, the planning module 2 is used to determine the surface to be cleaned of the first blade and its position information based on the three-dimensional model of the blade at the first parking moment, and to plan the cleaning trajectory of the first blade surface to be cleaned online. The cleaning module 3 is used to control the cleaning robot to clean the first blade surface to be cleaned according to the cleaning trajectory of the first blade surface to be cleaned. After the cleaning of the first blade surface to be cleaned is completed, the parking planning module is used to determine the second blade surface to be cleaned and its position information based on the preset correspondence between the surface to be cleaned and the blade position, and the preset blade cleaning process sequence. Specifically, the preset correspondence between the surface to be cleaned and the blade position is shown in Table 1, and the preset blade cleaning process sequence is shoulder → convex surface or concave surface → bottom. According to the position information of the second blade surface to be cleaned, the blade and the three-dimensional model of the blade are controlled to run to the position corresponding to the position information of the second blade surface to be cleaned. The cleaning trajectory of the second blade surface to be cleaned is planned online according to the second blade surface to be cleaned and the blade three-dimensional model; and the cleaning robot is controlled to clean the second blade surface to be cleaned according to the cleaning trajectory.

[0132] For example, based on the three-dimensional model of the blade at the first stop, the position information corresponding to the first stop and the first blade surface to be cleaned are determined. This position information is then compared with the revolution angle, solid blade pose 6dof, and hollow blade pose 6dof in Table 1 (traversal search). This determines the parking station number corresponding to the first stop and verifies the first blade surface to be cleaned. The second blade surface to be cleaned is determined based on the pre-set blade cleaning process sequence. If the first blade surface to be cleaned is a shoulder, the pre-set blade cleaning process sequence indicates whether the second blade surface to be cleaned is convex or concave. The pre-set correspondence between the surface to be cleaned and the blade position (Table 1) indicates that the parking station number corresponding to the second blade surface to be cleaned is "2." Based on parking station number "2," the corresponding position information for the second blade surface to be cleaned is obtained: the revolution angle, solid blade pose 6dof, and hollow blade pose 6dof. Based on the position information corresponding to the second blade surface to be cleaned, the blade and the three-dimensional blade model are controlled to move to the positions corresponding to the position information of the second blade surface to be cleaned. The trajectory planning module then plans the cleaning trajectory of the second blade surface to be cleaned online. The cleaning module then controls the cleaning robot to clean the second blade surface according to the cleaning trajectory. This cycle is repeated through multiple judgments to complete the cleaning of different blade surfaces.

[0133] Example 2

[0134] like Figure 9As shown, this embodiment provides an automatic cleaning method for the blade 301 of a vertical mixer, which is applied to the automatic cleaning system for the blade of a vertical mixer as described in Example 1. The server can obtain the actual posture information of the blade 301 through the device and feed it back to the control device 400; according to the actual posture information of the blade 301, the three-dimensional model of the blade 301 is adjusted from the basic position to the actual position corresponding to the actual posture information of the blade 301, and the cleaning trajectory is planned online based on the three-dimensional model of the blade 301 at the actual position; the cleaning robot 100 is controlled to clean the blade 301 according to the cleaning trajectory. The present application identifies the actual posture of the blade 301 to be cleaned, and adjusts the three-dimensional model of the blade 301 from the basic position to the actual position according to the actual posture of the blade 301, and plans the cleaning trajectory online based on the three-dimensional model of the blade 301 at the actual position, and uses the cleaning robot 100 to clean the blade 301 according to the cleaning trajectory. During the cleaning process of the blade 301, there is no need for personnel to enter the mixing workshop to manually clean the materials attached to the blade 301, which improves the safety of personnel and the safety of equipment use. This embodiment uses a three-dimensional model to plan the cleaning trajectory online, and there is no need to adjust the position of the blade 301, thereby improving the cleaning efficiency. Among them, the device end can be but is not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server end can be implemented with an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments.

[0135] Methods include:

[0136] S100: Acquire actual position information of the blade 301 and feed it back to the control device 400;

[0137] S200: According to the actual position and posture information of the blade 301, the three-dimensional model of the blade 301 is adjusted from the basic position to the actual position corresponding to the actual position and posture information of the blade 301, and a cleaning trajectory is planned online according to the three-dimensional model of the blade 301 at the actual position;

[0138] S300: Control the cleaning robot 100 to clean the blade 301 according to the cleaning trajectory.

[0139] Preferably, before obtaining the actual position information of the blade 301 in step S100, the method includes:

[0140] S01: Establish a coordinate system with the center of the revolution circle of the blade 301 as the origin, and create a three-dimensional model of the blade 301;

[0141] S02: Setting operating parameters of the three-dimensional model of the blade 301, simulating and obtaining three-dimensional digital-analog information of the blade 301 at a preset revolution angle within the revolution period, and obtaining a three-dimensional digital-analog information data set of the blade 301.

[0142] Specifically, in this embodiment, the three-dimensional models of the blade 301 and the cleaning robot 100 are drawn at a 1:1 ratio, and a coordinate system is established with the center of the revolution of the blade 301 as the origin. Thus, the positional relationship and spatial coordinates between the cleaning robot 100 and the blade 301 can be determined. The three-dimensional operating parameters of the blade 301 include the solid blade revolution speed, the solid blade rotation speed, the solid blade revolution trajectory, the solid blade rotation trajectory, the hollow blade revolution speed, the hollow blade rotation speed, the hollow blade revolution trajectory, and the hollow blade rotation trajectory. Figure 11 According to the three-dimensional model of the blade 301 and the three-dimensional operating parameters of the blade 301, in the 360-degree rotation range of the blade 301, the three-dimensional digital model information of the blade 301 at each revolution angle within the revolution period can be calculated based on the dynamic forward modeling. The corresponding three-dimensional digital model information data set of blade 301 is , the orbital angle of the hollow blade The corresponding three-dimensional digital model information data set of blade 301 is , n =1……360.

[0143] The preferred step S200 obtains three-dimensional digital model information of the blade 301 corresponding to the actual posture information of the blade 301 according to the actual posture information of the blade 301, and plans a cleaning trajectory online according to the corresponding three-dimensional digital model information of the blade 301, including:

[0144] S210: Matching the actual position and posture information of the blade 301 with the three-dimensional digital model information data set of the blade 301 to obtain the three-dimensional digital model information of the blade 301 that matches the actual position and posture information of the blade 301;

[0145] S220: Calculating an optimal deflection and translation of the three-dimensional model of the blade 301 from the base position to the actual position based on the three-dimensional digital model information of the blade 301 that matches the actual position information of the blade 301;

[0146] S230: adjusting the three-dimensional model of the blade 301 from the basic position to the actual position according to the optimal deflection and translation amount;

[0147] S240: Slice the three-dimensional model of the blade 301 at the actual position, generate a normal vector for each slice, and generate a cleaning trajectory online according to the normal vector for each slice.

[0148] Specifically, the optimal deflection and translation in step S220 include the optimal deflection and translation of the blade 301 for self-rotation and the optimal deflection and translation of the blade 301 for revolution. It is understandable that the optimal deflection and translation of the blade 301 for self-rotation include the optimal deflection and translation of the hollow blade for self-rotation and the optimal deflection and translation of the solid blade for self-rotation, and the optimal deflection and translation of the blade 301 for revolution include the optimal deflection and translation of the hollow blade for revolution and the optimal deflection and translation of the solid blade for co-rotation. When the three-dimensional digital-analog information data set of the blade 301 is The three-dimensional digital model information of the solid blade that matches the actual position information of the solid blade is obtained, and then the optimal deflection and translation measurement of the solid blade rotation and the optimal deflection and translation measurement of the solid blade revolution are obtained; in the three-dimensional digital model information data set of blade 301, the three-dimensional digital model information of the solid blade is obtained. The three-dimensional digital model information of the hollow blade that matches the actual position information of the hollow blade is obtained, and then the optimal deflection and translation measurement of the hollow blade's rotation and the optimal deflection and translation measurement of the hollow blade's revolution are obtained.

[0149] Specifically, this embodiment draws a 1:1 three-dimensional model of the blade 301 and a three-dimensional model of the cleaning robot 100, meshes the three-dimensional model, and imports it into the control device 400 through the input device 500. The three-dimensional digital model information of the blade 301 at a preset revolution angle within the revolution period is simulated to form a three-dimensional digital model information data set of the blade 301. The structured light 3D scanner 201 of the posture recognition device 200 obtains the actual posture three-dimensional point cloud image data of the blade 301, and the three-dimensional point cloud image data is provided to the back-end platform 202. The back-end platform 202 parses the three-dimensional point cloud image data to identify the spatial coordinate information of the blade 301 to be cleaned and provides it to the control device 400. The spatial coordinate information of the blade 301 to be cleaned is then converted into a coordinate system with the center of the revolution circle of the three-dimensional model of the blade 301 as the origin. Since the positional relationship and spatial coordinates between the cleaning robot 100 and the blade 301 have been determined, the three-dimensional digital-analog information of the blade 301 at each revolution angle within the revolution period includes the spatial coordinate information of the three-dimensional model of the blade 301 at each revolution angle. The spatial coordinate information of the blade 301 to be cleaned and the spatial coordinate information of the three-dimensional model of the blade 301 at each revolution angle are matched to obtain the optimal deflection and translation.

[0150] Specifically, the actual position and posture information of blade 301 is matched against the three-dimensional digital model information dataset of blade 301 to obtain three-dimensional digital model information of blade 301 that matches the actual position and posture information of blade 301. Based on the three-dimensional digital model information of blade 301 that matches the actual position and posture information of blade 301, the optimal deflection and translation of the three-dimensional model of blade 301 from the base position to the actual position are calculated. Based on the optimal deflection and translation, the three-dimensional model of blade 301 is adjusted from the base position to the actual position, completing the motion modeling of blade 301. A topological analysis is performed on the three-dimensional model of blade 301, and a path line is obtained by slicing. Then, a normal vector is generated for the slice. The sliced ​​path line can be converted into a cleaning trajectory, thereby obtaining the running trajectory (motion point information) of cleaning robot 100 corresponding to cleaning the curved surface of each blade 301 at each position. Based on the running trajectory, cleaning robot 100 sequentially cleans blade 301.

[0151] The method further includes: cleaning the surface to be cleaned of the first blade at a first stop time, and cleaning the surface to be cleaned of the second blade at a second stop time, wherein the first stop time and the second stop time are adjacent to each other. The first stop time and the second stop time correspond to two adjacent stops, and do not specifically refer to the number of stops. That is, in the adjacent stops on both sides, the first stop time is the previous stop time, and the second stop time is the next stop time.

[0152] Specifically, in one embodiment, based on the three-dimensional model of the blade at the first parking moment, a first blade surface to be cleaned and its position information are determined, and a cleaning trajectory for the first blade surface to be cleaned is planned online; the cleaning robot is controlled to clean the first blade surface to be cleaned according to the cleaning trajectory of the first blade surface to be cleaned. After cleaning the first blade surface to be cleaned, the method further includes: determining a second blade surface to be cleaned and its position information based on the first blade surface to be cleaned, based on a preset correspondence between the surface to be cleaned and the blade position, and a preset blade cleaning process sequence. Specifically, the preset correspondence between the surface to be cleaned and the blade position is shown in Table 1, and the preset blade cleaning process sequence is shoulder → convex surface or concave surface → bottom. Based on the position information of the second blade surface to be cleaned, the blade and the three-dimensional model of the blade are controlled to move to a position corresponding to the position information of the second blade surface to be cleaned. A cleaning trajectory for the second blade surface to be cleaned is planned online based on the second blade surface to be cleaned and the three-dimensional model of the blade; and the cleaning robot is controlled to clean the second blade surface to be cleaned according to the cleaning trajectory of the second blade surface to be cleaned.

[0153] For example, based on the three-dimensional model of the blade at the first stop, the position information corresponding to the first stop and the first blade surface to be cleaned are determined. The position information corresponding to the first stop is compared with the revolution angle, solid blade pose 6dof, and hollow blade pose 6dof in Table 1 to determine the parking station number corresponding to the first stop, and the first blade surface to be cleaned is verified. The second blade surface to be cleaned is determined based on the pre-set blade cleaning process sequence. If the first blade surface to be cleaned is a shoulder, the pre-set blade cleaning process sequence indicates whether the second blade surface to be cleaned is convex or concave. The pre-set correspondence between the surface to be cleaned and the blade position (Table 1) indicates that the parking station number corresponding to the second blade surface to be cleaned is "2". Based on the parking station number "2", the position information corresponding to the second blade surface to be cleaned is obtained: the revolution angle, solid blade pose 6dof, and hollow blade pose 6dof. Based on the position information corresponding to the second blade surface to be cleaned, the blade and the three-dimensional blade model are controlled to move to the positions corresponding to the position information of the second blade surface to be cleaned. The trajectory planning module then plans the cleaning trajectory of the second blade surface to be cleaned online. The cleaning module then controls the cleaning robot to clean the second blade surface according to the cleaning trajectory. This cycle is repeated through multiple judgments to complete the cleaning of different blade surfaces.

[0154] Example 3

[0155] like Figure 9As shown, this embodiment provides an automatic cleaning method for the blade 301 of a vertical mixer, which is applied to the automatic cleaning system for the blade of a vertical mixer as described in Example 2. The server can obtain the actual posture information of the blade 301 through the device and feed it back to the control device 400; according to the actual posture information of the blade 301, the three-dimensional model of the blade 301 is adjusted from the basic position to the actual position corresponding to the actual posture information of the blade 301, and the cleaning trajectory is planned online based on the three-dimensional model of the blade 301 at the actual position; the cleaning robot 100 is controlled to clean the blade 301 according to the cleaning trajectory. The present application identifies the actual posture of the blade 301 to be cleaned, and adjusts the three-dimensional model of the blade 301 from the basic position to the actual position according to the actual posture of the blade 301, and plans the cleaning trajectory online based on the three-dimensional model of the blade 301 at the actual position, and uses the cleaning robot 100 to clean the blade 301 according to the cleaning trajectory. During the cleaning process of the blade 301, there is no need for personnel to enter the mixing workshop to manually clean the materials attached to the blade 301, which improves the safety of personnel and the safety of equipment use. This implementation uses a three-dimensional model to plan the cleaning trajectory online, without adjusting the position of the blade 301, thereby improving the cleaning efficiency. Among them, the device end can be but is not limited to various personal computers, laptops, smart phones, tablets and portable wearable devices. The server end can be implemented with an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments. The method includes:

[0156] S100: Acquire actual position information of the blade 301 and feed it back to the control device 400;

[0157] S200: According to the actual position and posture information of the blade 301, the three-dimensional model of the blade 301 is adjusted from the basic position to the actual position corresponding to the actual position and posture information of the blade 301, and a cleaning trajectory is planned online according to the three-dimensional model of the blade 301 at the actual position;

[0158] S300: Control the cleaning robot 100 to clean the blade 301 according to the cleaning trajectory.

[0159] Preferably, before obtaining the actual position information of the blade 301 in step S100, the method includes:

[0160] S01: Establish a coordinate system with the center of the revolution circle of the blade 301 as the origin, and create a three-dimensional model of the blade 301;

[0161] S02: obtaining a point cloud of the basic position of the blade at the basic position according to the three-dimensional model of the blade 301;

[0162] S03: performing point cloud registration on the actual blade pose point cloud and the basic blade pose point cloud to obtain the predicted deflection and translation of the blade 3D model from the basic position to the actual position;

[0163] S04: Based on the preset motion trajectory of the blade drive shaft axis, and according to the predicted deflection and translation, the posture of the blade three-dimensional model adjusted to the actual position is corrected to obtain the optimal deflection and translation.

[0164] Specifically, the actual blade position information includes a point cloud of the actual blade position, which is obtained by scanning with a structured light 3D scanner.

[0165] Specifically, in another embodiment, the basic pose point cloud of the blade at the basic position can be obtained by scanning with a structured light 3D scanner.

[0166] Specifically, such as Figure 13 As shown, step S03 includes point cloud preprocessing of the actual blade pose point cloud, including removing large point clouds outside the blade motion range through through filtering, and removing outliers through statistical filtering to obtain high-quality point clouds.

[0167] After obtaining a high-quality point cloud, step S03 further includes:

[0168] Extracting point cloud feature points from the actual position point cloud of the blade based on a preset feature point extraction model;

[0169] Calculating a feature histogram of the point cloud feature points based on a preset feature extraction model;

[0170] Based on a preset outlier removal model, outliers are removed according to the feature histogram to obtain the initial posture of the blade at the actual position.

[0171] Specifically, in this embodiment, the feature point extraction model is the ISS (Intrinsic Shape Signature) algorithm, which extracts the point cloud feature points in the actual position point cloud of the blade. The feature extraction model is the FPFH (Intrinsic Shape Signature) algorithm, which calculates the feature histogram of the point cloud feature points and performs corresponding point matching based on the feature histogram of the point cloud feature points. The abnormal point removal model is the RANSAC (Random Sample Consensus) algorithm, which eliminates erroneous point pair relationships. The processing process of the initial posture of the blade at the actual position solves the problem of external points in the sample and effectively eliminates some points that do not conform to the optimal parameter model, such as noise, invalid points, etc. After eliminating the erroneous point pair relationships, matching can be performed based on the remaining point cloud relationships.

[0172] After obtaining a high-quality point cloud, step S03 further includes:

[0173] The actual blade pose point cloud and the basic blade pose point cloud are divided into grids, and the number of points in each grid is counted. Based on the number of points in each grid, the mean and covariance of each grid are calculated to construct the Gaussian distribution of the points in each grid.

[0174] According to the initial posture of the blade at the actual position and the Gaussian distribution of the points in each grid, the initial values ​​of the deflection and translation between the actual posture point cloud of the blade and the basic posture point cloud of the blade are calculated;

[0175] According to the initial values ​​of the offset and translation, the joint probability that the actual blade pose point cloud falls into the basic blade pose point cloud is calculated;

[0176] The predicted deflection and translation are calculated based on the maximum value of the joint probabilities.

[0177] Step S03 applies the NDT (normal distributions transform) algorithm, which converts a point cloud dataset within a 3D voxel into a continuously differentiable probability distribution function for point cloud matching. The NDT algorithm divides the blade's actual pose point cloud and the blade's basic pose point cloud into grids, forming uniform, regular, and fixed-size 3D grids. Within each grid, individual point cloud sets are formed. By applying a normal probability distribution to the point cloud sets within each grid, the blade's actual pose point cloud and the blade's basic pose point cloud are calculated against each other repeatedly, ultimately achieving a matching result. Throughout the algorithm, the normal distribution probability density within the grid only needs to be calculated once, eliminating the need for corresponding calculations and matching operations for each point. Therefore, the algorithm is computationally lightweight and highly efficient.

[0178] Specifically, since the blade surface is "contaminated" by material during actual application, it is necessary to use the axial motion trajectory of the blade drive shaft to correct the alignment result, so as to ensure the accuracy of the output blade posture. Figure 11 The measured values ​​of the axis coordinates of the hollow propeller (the axis motion trajectory of the hollow blade drive shaft) and the measured values ​​of the axis coordinates of the solid propeller (the axis motion trajectory of the solid blade drive shaft) are shown in FIG.

[0179] Specifically, step S04 includes: calculating and obtaining the estimated axis center position coordinates of the three-dimensional model of the blade adjusted to the actual position according to the predicted deflection amount and translation amount;

[0180] When the estimated axis center position coordinates of the blade three-dimensional model adjusted to the actual position do not overlap with the axis center motion trajectory of the blade transmission shaft, determining the matching point coordinates of the estimated axis center position coordinates of the blade three-dimensional model adjusted to the actual position in the axis center motion trajectory of the blade transmission shaft;

[0181] Calculate the optimal deflection and translation amounts according to the matching point coordinates.

[0182] Specifically, in this embodiment, the current blade position is clearly determined by the output optimal deflection R and translation T. The calibration results of the structured light 3D scanner 201 can then be transferred to the blade's orbital coordinate system and displayed in 3D, allowing for intuitive verification of the correctness of the registration. Furthermore, based on the calibration results of the structured light 3D scanner 201 and the cleaning robot 100 coordinate system, the blade position registered using the structured light 3D scanner 201 can be transferred to the cleaning robot 100 coordinate system for use.

[0183] The method further includes: cleaning the surface to be cleaned of the first blade at a first stop time, and cleaning the surface to be cleaned of the second blade at a second stop time, wherein the first stop time and the second stop time are adjacent to each other. The first stop time and the second stop time correspond to two adjacent stops, and do not specifically refer to the number of stops. That is, in the adjacent stops on both sides, the first stop time is the previous stop time, and the second stop time is the next stop time.

[0184] Specifically, in one embodiment, based on the three-dimensional model of the blade at the first parking moment, a first blade surface to be cleaned and its position information are determined, and a cleaning trajectory for the first blade surface to be cleaned is planned online; the cleaning robot is controlled to clean the first blade surface to be cleaned according to the cleaning trajectory of the first blade surface to be cleaned. After cleaning the first blade surface to be cleaned, the method further includes: determining a second blade surface to be cleaned and its position information based on the first blade surface to be cleaned, based on a preset correspondence between the surface to be cleaned and the blade position, and a preset blade cleaning process sequence. Specifically, the preset correspondence between the surface to be cleaned and the blade position is shown in Table 1, and the preset blade cleaning process sequence is shoulder → convex surface or concave surface → bottom. Based on the position information of the second blade surface to be cleaned, the blade and the three-dimensional model of the blade are controlled to move to a position corresponding to the position information of the second blade surface to be cleaned. A cleaning trajectory for the second blade surface to be cleaned is planned online based on the second blade surface to be cleaned and the three-dimensional model of the blade; and the cleaning robot is controlled to clean the second blade surface to be cleaned according to the cleaning trajectory of the second blade surface to be cleaned.

[0185] For example, based on the three-dimensional model of the blade at the first stop, the position information corresponding to the first stop and the first blade surface to be cleaned are determined. The position information corresponding to the first stop is compared with the revolution angle, solid blade pose 6dof, and hollow blade pose 6dof in Table 1 to determine the parking station number corresponding to the first stop, and the first blade surface to be cleaned is verified. The second blade surface to be cleaned is determined based on the pre-set blade cleaning process sequence. If the first blade surface to be cleaned is a shoulder, the pre-set blade cleaning process sequence indicates whether the second blade surface to be cleaned is convex or concave. The pre-set correspondence between the surface to be cleaned and the blade position (Table 1) indicates that the parking station number corresponding to the second blade surface to be cleaned is "2". Based on the parking station number "2", the position information corresponding to the second blade surface to be cleaned is obtained: the revolution angle, solid blade pose 6dof, and hollow blade pose 6dof. Based on the position information corresponding to the second blade surface to be cleaned, the blade and the three-dimensional blade model are controlled to move to the positions corresponding to the position information of the second blade surface to be cleaned. The trajectory planning module then plans the cleaning trajectory of the second blade surface to be cleaned online. The cleaning module then controls the cleaning robot to clean the second blade surface according to the cleaning trajectory. This cycle is repeated through multiple judgments to complete the cleaning of different blade surfaces.

[0186] This invention provides an automatic cleaning method for vertical mixer blades. A position recognition device identifies the actual position of the blade to be cleaned. Based on the actual position, the blade's three-dimensional model is adjusted from a base position to its actual position. A cleaning trajectory is then planned online based on the blade's three-dimensional model at the actual position. A cleaning robot then cleans the blade along the cleaning trajectory. This process eliminates the need for personnel to enter the mixing room to manually clean material adhering to the blade, improving both personnel and equipment safety.

[0187] Example 4

[0188] This embodiment also provides a vertical mixer blade automatic cleaning device, which corresponds to the vertical mixer blade automatic cleaning method in the second and third embodiments, and can refer to the control device 400 in the first embodiment. Figure 14 As shown, the vertical mixer blade automatic cleaning device includes an acquisition module 1, a planning module 2, and a cleaning module 3. The functional modules are described in detail as follows:

[0189] Acquisition module 1 is used to obtain the actual position information of the blade;

[0190] Planning module 2, configured to adjust the three-dimensional model of the blade from a base position to an actual position corresponding to the actual position of the blade according to the actual position of the blade, and to plan a cleaning trajectory online according to the three-dimensional model of the blade at the actual position;

[0191] The cleaning module 3 is used to control the cleaning robot to clean the blades according to the cleaning trajectory.

[0192] The acquisition module 1 is also used to obtain the basic posture point cloud of the blade at the basic position.

[0193] Planning module 2, including a simulation module, a first matching module and a first calculation module;

[0194] The simulation module is used to simulate and obtain a three-dimensional digital-analog information data set of the blade according to the three-dimensional model of the blade and preset blade operating parameters;

[0195] The first matching module is used to match the actual blade posture information with the blade three-dimensional digital-analog information data set to obtain the blade three-dimensional digital-analog information that matches the actual blade posture information;

[0196] The first calculation module is used to calculate the optimal deflection and translation of the blade three-dimensional model from the basic position to the actual position based on the blade three-dimensional digital model information matching the blade actual posture information.

[0197] The planning module 2 is further configured to adjust the three-dimensional model of the blade from a basic position to an actual position according to the optimal deflection and translation amounts;

[0198] The three-dimensional model of the propeller blade at the actual position is sliced, and a normal vector of each slice is generated. A cleaning trajectory is generated online according to the normal vector of each slice.

[0199] Planning module 2, further comprising a second matching module, a second calculation module and a correction module;

[0200] The second matching module is used to perform point cloud registration between the actual blade posture information and the basic blade posture information;

[0201] The second calculation module is used to calculate the predicted deflection and predicted translation of the three-dimensional model of the blade from the basic position to the actual position;

[0202] The correction module is used to correct the posture of the blade three-dimensional model adjusted to the actual position based on the preset blade transmission shaft axis motion trajectory and the predicted deflection amount and the predicted translation amount.

[0203] The second matching module is further configured to extract point cloud feature points from the actual blade pose point cloud based on a preset feature point extraction model;

[0204] Calculating a feature histogram of the point cloud feature points based on a preset feature extraction model;

[0205] Based on a preset outlier removal model, outliers are removed according to the feature histogram to obtain the initial posture of the blade at the actual position.

[0206] The second calculation module is further configured to calculate, based on the predicted deflection and translation, the estimated axis center position coordinates of the three-dimensional model of the blade adjusted to the actual position;

[0207] When the estimated axis center position coordinates of the blade three-dimensional model adjusted to the actual position do not overlap with the axis center motion trajectory of the blade transmission shaft, determining the matching point coordinates of the estimated axis center position coordinates of the blade three-dimensional model adjusted to the actual position in the axis center motion trajectory of the blade transmission shaft;

[0208] Calculate the optimal deflection and translation amounts according to the matching point coordinates.

[0209] The automatic cleaning device for the blades of a vertical mixer also includes a parking planning module, which is used to determine the second blade surface to be cleaned and its position information according to the first blade surface to be cleaned based on the preset correspondence between the surface to be cleaned and the blade position, and the preset blade cleaning process sequence; according to the position information of the second blade surface to be cleaned, control the blade and the blade three-dimensional model to run to the position corresponding to the position information of the second blade surface to be cleaned.

[0210] Planning module 2 is also used to control the blade and the blade three-dimensional model to run to the position corresponding to the position information of the second blade surface to be cleaned according to the position information of the second blade surface to be cleaned; and to plan the cleaning trajectory of the second blade surface to be cleaned online according to the second blade surface to be cleaned and the blade three-dimensional model.

[0211] The cleaning module 3 is further used to control the cleaning robot to clean the surface to be cleaned of the second blade according to the cleaning trajectory of the surface to be cleaned of the second blade.

[0212] The present invention provides an automatic cleaning device for vertical mixer blades. A position recognition device identifies the actual position of the blade to be cleaned. Based on the actual position, the blade's three-dimensional model is adjusted from a base position to its actual position. A cleaning trajectory is then planned online based on the blade's three-dimensional model at the actual position. A cleaning robot then cleans the blade along the cleaning trajectory. This process eliminates the need for personnel to enter the mixing room to manually clean material adhering to the blade, improving both personnel and equipment safety.

[0213] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 15As shown. The computer device includes a processor, memory, network interface and database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external device end via a network connection. When the computer program is executed by the processor, it realizes the functions or steps of the service end side of a vertical mixer blade automatic cleaning method.

[0214] In one embodiment, a computer device is provided. The computer device may be a device end, and its internal structure diagram may be as follows: Figure 16 As shown. The computer device includes a processor, memory, network interface, display screen, and input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a method for automatically cleaning the blades of a vertical mixer on the device side.

[0215] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed:

[0216] Get the actual position information of the blade;

[0217] According to the actual blade posture information, the three-dimensional model of the blade is adjusted from a basic position to an actual position corresponding to the actual blade posture information, and a cleaning trajectory is planned online based on the three-dimensional model of the blade at the actual position;

[0218] The cleaning robot is controlled to clean the blades according to the cleaning trajectory.

[0219] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0220] Get the actual position information of the blade;

[0221] According to the actual blade posture information, the three-dimensional model of the blade is adjusted from a basic position to an actual position corresponding to the actual blade posture information, and a cleaning trajectory is planned online based on the three-dimensional model of the blade at the actual position;

[0222] The cleaning robot is controlled to clean the blades according to the cleaning trajectory.

[0223] It should be noted that the above functions or steps that can be implemented by computer-readable storage media or computer devices can be found in the relevant descriptions of the server side and the device side in the aforementioned method embodiments. To avoid repetition, they will not be described one by one here.

[0224] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0225] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0226] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

[0227] In the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The term "plurality" refers to two or more than two, unless otherwise clearly defined.

[0228] Other embodiments of the present invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as illustrative only.

[0229] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for automatically cleaning blades of a vertical mixer, characterized in that: The method comprises: Acquire the actual propeller position information, which is identified by a position recognition device. The cleaning robot is located on one side of the mixer, and the position recognition device is located on the other side of the mixer. The position recognition device includes a structured light 3D scanner. The structured light 3D scanner can identify the propeller position in any parking space with a 6Dof (degree of freedom) accuracy within a working distance of more than 1.8 meters. The structured light 3D scanner uses blue natural light for photo recognition. According to the actual blade posture information, the three-dimensional model of the blade is adjusted from a basic position to an actual position corresponding to the actual blade posture information, and a cleaning trajectory is planned online based on the three-dimensional model of the blade at the actual position; Controlling the cleaning robot to clean the blades according to the cleaning trajectory; When the blades are automatically cleaned, the first blade surface to be cleaned is cleaned at the first stop time, and the second blade surface to be cleaned is cleaned at the second stop time, and the first stop time and the second stop time are adjacent to each other; The method further comprises: determining a first blade position and a first blade surface to be cleaned corresponding to the first parking moment based on the three-dimensional model of the blade at the first parking moment; Determining the second blade surface to be cleaned and its position information based on the first blade position and the first blade surface to be cleaned corresponding to the first parking moment, based on a preset correspondence between the surface to be cleaned and the blade position, and a preset blade cleaning process sequence; According to the position information of the second blade surface to be cleaned, controlling the blade and the blade three-dimensional model to move to a position corresponding to the position information of the second blade surface to be cleaned; Online planning of a cleaning trajectory of the second blade surface to be cleaned based on the second blade surface to be cleaned and the three-dimensional model of the blade; controlling the cleaning robot to clean the surface to be cleaned of the second blade according to the cleaning trajectory of the surface to be cleaned of the second blade; The actual blade position information includes a point cloud of the actual blade position; The adjusting, according to the actual blade posture information, the blade three-dimensional model from a basic position to an actual position corresponding to the actual blade posture information includes: Obtaining a basic position point cloud of the blade at a basic position according to the three-dimensional model of the blade; Performing point cloud registration on the actual blade pose point cloud and the basic blade pose point cloud to obtain a predicted deflection and translation of the blade three-dimensional model from the basic position to the actual position; Based on the preset motion trajectory of the blade drive shaft axis, and according to the predicted deflection and translation, the posture of the blade three-dimensional model adjusted to the actual position is corrected to obtain the optimal deflection and translation; The method of correcting the posture of the three-dimensional blade model adjusted to the actual position based on the preset blade transmission shaft axis motion trajectory and the predicted deflection and translation to obtain the optimal deflection and translation includes: Calculating the estimated axis center position coordinates of the three-dimensional model of the blade adjusted to the actual position based on the predicted deflection and translation; When the estimated axis center position coordinates of the blade three-dimensional model adjusted to the actual position do not overlap with the axis center motion trajectory of the blade transmission shaft, determining the matching point coordinates of the estimated axis center position coordinates of the blade three-dimensional model adjusted to the actual position in the axis center motion trajectory of the blade transmission shaft; Calculate the optimal deflection and translation amounts according to the matching point coordinates.

2. The automatic cleaning method for vertical mixer blades according to claim 1, characterized in that: Before obtaining the actual blade posture information, the method includes: A coordinate system is established with the center of the blade's revolution as the origin, and a three-dimensional model of the blade is created; The operating parameters of the three-dimensional model of the blade are set, and the three-dimensional digital-analog information of the blade at a preset revolution angle within the revolution period is obtained by simulation to obtain a three-dimensional digital-analog information data set of the blade.

3. The automatic cleaning method for vertical mixer blades according to claim 2, characterized in that: The operating parameters of the blade three-dimensional model include the solid blade revolution speed, the solid blade rotation speed, the solid blade revolution trajectory, the solid blade rotation trajectory, the hollow blade revolution speed, the hollow blade rotation speed, the hollow blade revolution trajectory, and the hollow blade rotation trajectory.

4. The automatic cleaning method for vertical mixer blades according to claim 2, characterized in that: The method further comprises adjusting the three-dimensional blade model from a base position to an actual position corresponding to the actual blade position information according to the actual blade position information, and online planning a cleaning trajectory according to the three-dimensional blade model at the actual position, including: Matching the actual blade pose information with the blade three-dimensional digital model information data set to obtain blade three-dimensional digital model information that matches the actual blade pose information; Based on the three-dimensional digital model of the blade that matches the actual blade position information, the optimal deflection and translation of the three-dimensional blade model from the basic position to the actual position is calculated; adjusting the three-dimensional model of the blade from a basic position to an actual position according to the optimal deflection and translation; The three-dimensional model of the propeller blade at the actual position is sliced, and a normal vector of each slice is generated. A cleaning trajectory is generated online according to the normal vector of each slice.

5. An automatic cleaning device for vertical mixer blades, characterized in that: include: an acquisition module for acquiring actual position information of the blade, the actual position information of the blade being identified by a position recognition device; a cleaning robot being located on one side of the mixer, and the position recognition device being located on the other side of the mixer; the cleaning robot comprising a scraper, a plurality of cleaning structures being provided on one scraper, a cleaning structure being matched with a surface to be cleaned of the blade, and the plurality of cleaning structures being used to clean the plurality of surfaces to be cleaned of the blade; the plurality of cleaning structures comprising an arc-shaped concave surface structure, an arc-shaped convex surface structure and a flat surface structure, the arc-shaped concave surface structure being used to clean the convex surface to be cleaned of the blade, the arc-shaped convex surface structure being used to clean the concave surface to be cleaned of the blade, and the flat surface structure being used to clean the flat surface to be cleaned of the blade; the position recognition device comprising a structured light 3D scanner, the structured light 3D scanner being able to identify the 6dof posture of the blade in any parking space with millimeter-level recognition accuracy within a working distance of more than 1.8 meters with a large viewing distance, and the actual position information of the blade comprising a point cloud of the actual position of the blade; a planning module, configured to adjust the three-dimensional model of the blade from a base position to an actual position corresponding to the actual position of the blade according to the actual position of the blade, and to plan a cleaning trajectory online according to the three-dimensional model of the blade at the actual position; A cleaning module, configured to control the cleaning robot to clean the blades according to the cleaning trajectory; A planning module, configured to determine a first blade position and a surface to be cleaned of the first blade corresponding to the first parking moment based on the three-dimensional model of the blade at the first parking moment; a parking planning module, configured to determine information on a second blade surface to be cleaned and its position based on the first blade position and the first blade surface to be cleaned corresponding to the first parking moment, a preset correspondence between the surface to be cleaned and the blade position, and a preset blade cleaning process sequence; a planning module, configured to control the blade and the three-dimensional model of the blade to move to a position corresponding to the position information of the second blade surface to be cleaned based on the position information of the second blade surface to be cleaned; and to online plan a cleaning trajectory of the second blade surface to be cleaned based on the second blade surface to be cleaned and the three-dimensional model of the blade; a cleaning module, configured to control the cleaning robot to clean the surface to be cleaned of the second blade according to a cleaning trajectory of the surface to be cleaned of the second blade; The acquisition module is used to obtain a basic position point cloud of the blade at a basic position according to the three-dimensional model of the blade; The second matching module is used to perform point cloud registration between the actual blade pose point cloud and the basic blade pose point cloud; The second calculation module is used to calculate and obtain the predicted deflection and translation of the three-dimensional model of the blade from the basic position to the actual position; The correction module is used to correct the posture of the blade three-dimensional model adjusted to the actual position based on the preset blade drive shaft axis motion trajectory and the predicted deflection and translation to obtain the optimal deflection and translation; The second calculation module is also used to calculate the estimated axial position coordinates of the blade three-dimensional model adjusted to the actual position based on the predicted deflection and translation. When the estimated axial position coordinates of the blade three-dimensional model adjusted to the actual position do not overlap with the axial motion trajectory of the blade transmission shaft, the matching point coordinates of the estimated axial position coordinates of the blade three-dimensional model adjusted to the actual position are determined in the axial motion trajectory of the blade transmission shaft, and the optimal deflection and translation are calculated based on the matching point coordinates.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for automatically cleaning the blades of a vertical mixer according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for automatically cleaning the blades of a vertical mixer according to any one of claims 1 to 4 are implemented.

Images