Brick surface targeted slurry laying system and method based on robot

Abstract

The invention belongs to the technical field of coke oven masonry, and particularly discloses a brick surface targeted slurry laying system and method based on a robot. Comprising a robot; the slurry laying module comprises a feeding unit and a discharging port connected with the feeding unit, the feeding unit is used for conveying slurry with specified pressure to the discharging port, and the discharging port is clamped at the tail end of a robot; the visual module is used for acquiring visual information of the robot, the discharge port and the coke oven masonry; and the control module is used for planning a motion track of the robot according to the visual information, calculating an angle error and a position error of the robot according to an expected pose of the discharge port, and compensating a joint angle and a joint position of the robot according to the angle error and the position error. And the robot drives the discharge port to perform accurate laying on the specified brick surface according to the preset motion trail. According to the invention, multi-face brick directional laying can be realized, and the slurry laying precision and efficiency are high.

Description

Technical Field

[0001] The present invention belongs to the technical field of coke oven masonry, and more particularly, relates to a robot-based brick surface targeted slurry laying system and method. Background Art

[0002] Coke ovens, crucial equipment for converting coal into coke, feature complex and diverse structures, diverse brick shapes, and significant variations in individual brick weight, ranging from a few kilograms to dozens of kilograms. Traditionally, coke oven masonry construction relied primarily on manual labor. However, with rising labor costs and increasing labor scarcity, the application of automation technology in coke oven construction has become increasingly urgent. Robotic arms are gradually entering the field of coke oven construction, aiming to improve efficiency and quality through mechanization.

[0003] However, Chinese patent CN117801831A discloses "a masonry machine, masonry method and system based on dual-arm collaboration", which has made significant progress in automated masonry, realized the simultaneous clamping of refractory bricks and mud spraying, and greatly improved the masonry efficiency, but there are still certain deficiencies in the precise control and uniformity of mud laying. In this solution, the second robotic arm is equipped with a refractory mud spraying gun head, which uses a second depth camera to identify the mud spraying position of the refractory bricks to be laid and perform mud spraying. However, in actual operation, factors such as mud flow, spraying uniformity, and the distance between the spraying gun head and the brick surface may affect the mud laying effect. Since this technology does not elaborate on the precise control method of mud flow and the strategy for uniform mud spraying on complex or curved structures, there may be a problem of uneven mud distribution. Especially when dealing with refractory bricks of different models and specifications, the determination of the starting and ending positions of spraying and the planning of the spraying path may have an adverse effect on the uniformity and accuracy of the slurry laying.

[0004] Furthermore, while the technical documentation does not elaborate on spraying details and technical parameters, it is reasonable to infer that in actual application, the design and use of the spray gun tip may be affected by various factors such as mud viscosity, spraying speed, and spraying distance, which in turn affects the uniform spraying of the mud. More importantly, the spraying process itself is difficult to completely avoid air inclusion, which may lead to uneven mud spraying and prone to splashing. Splashing not only affects the quality of masonry, such as causing weak bonds between refractory bricks, but also may pollute the surrounding environment, increasing the difficulty and cost of cleanup.

[0005] Due to the instantaneous nature and high force of spraying, this technology struggles to achieve the precise slurry application required by manual masonry. Automated spraying can be particularly challenging for complex masonry structures or curved surfaces, resulting in inconsistent quality. In contrast, manual masonry offers greater controllability and precision in slurry application. Masons can flexibly adjust the slurry application method and amount to suit specific needs, ensuring uniform and stable application. This flexibility is currently difficult for automated masonry robots to fully replace.

[0006] Therefore, how to balance the controllability and accuracy of mud spraying while pursuing automation and efficiency remains one of the key challenges facing current masonry robot technology. Summary of the Invention

[0007] In response to the above-mentioned defects or improvement needs of the prior art, the present invention provides a robot-based targeted slurry laying system and method for brick surfaces, which is mainly used for accurately laying slurry on the side and top surfaces of bricks (including the side surfaces on which the brick surfaces are fixed during the masonry process and the top surfaces on which bricks will be subsequently laid). The system combines a robot with a unique slurry conveying and laying device, and uses a three-dimensional vision system to accurately identify the position and status of the brick surface, providing precise guidance for the robot's slurry laying action. It also has an obstacle avoidance function to prevent the robot from colliding with other objects in a complex masonry environment. A support is provided at the end of the robot, on which a cylinder or electric cylinder that can provide pressure is fixed, and the slurry is conveyed and extruded through a connected tee and feed pipe. In addition, the system is also equipped with a walking axis and a vertical column that can be raised and lowered and rotated to expand the robot's operating range and operational flexibility. At the same time, the hopper set within the construction range and the internal spiral conveying device can achieve a continuous supply of slurry. The present invention can ensure the uniform distribution of mud on the brick surface, improve the accuracy and efficiency of mud laying, provide high-quality mud laying guarantee for masonry projects, avoid the problem of uneven mud distribution in traditional mud laying methods, and improve the overall masonry quality.

[0008] To achieve the above objectives, according to one aspect of the present invention, a robot-based targeted grouting system for brick surfaces is proposed, comprising:

[0009] robot;

[0010] A slurry laying module, comprising a feeding unit and a discharge port connected to the feeding unit, wherein the feeding unit is used to deliver slurry of a specified pressure to the discharge port, and the discharge port is clamped at the end of the robot;

[0011] Vision module, used to obtain visual information of the robot, discharge port and coke oven masonry;

[0012] The control module is used to plan the robot's motion trajectory based on the visual information, calculate the robot's angular error and position error based on the expected posture of the discharge port, and compensate the robot's joint angle and position based on the angular error and position error, so that the robot drives the discharge port to accurately lay the specified brick surface according to the predetermined motion trajectory.

[0013] As a further preference, the feeding unit includes a hopper for placing mud, and a first fluid diverter is provided at the bottom of the hopper. The first fluid diverter is connected to the discharge port through a feed pipe, and is used to divert the mud in the hopper to the feed pipe. A second fluid diverter is also provided between the feed pipe and the discharge port.

[0014] As a further preference, a spiral blade is provided between the first fluid diverter and the second fluid diverter, and the spiral blade is a shaftless spiral blade, which is arranged along the central axis of the feed pipe and is driven to rotate by a motor.

[0015] As a further preferred embodiment, a gradual buffer chamber is provided in the middle of the feed pipe, and the inner diameter of the gradual buffer chamber gradually decreases along the direction from the first fluid diverter to the second fluid diverter to adjust the flow rate and pressure of the mud and discharge the gas in the mud at the same time.

[0016] As a further preferred embodiment, a pre-pressure limiting valve is provided between the first fluid diverter and the spiral blade, and the pre-pressure limiting valve is used to open when the mud pressure is greater than a preset value to convey mud to the spiral blade.

[0017] As a further preferred embodiment, it also includes a walking drive module for driving the robot to walk, the walking drive module includes a horizontal walking unit and a lifting unit, the horizontal walking unit is used to drive the robot to move horizontally along the coke oven masonry, and the lifting unit is used to drive the robot to move vertically along the coke oven masonry.

[0018] As a further preferred embodiment, the horizontal walking unit includes a horizontal support member, a horizontal screw drive assembly provided on the horizontal support member, and a support plate is provided on the sliding nut of the horizontal screw drive assembly; the lifting unit is provided on the support plate, and the lifting unit includes a vertical support member and a vertical screw drive assembly provided on the vertical support member, and the sliding nut of the vertical screw drive assembly is fixedly connected to the robot.

[0019] According to another aspect of the present invention, a robot-based method for precise grouting on brick surfaces is provided, comprising the following steps:

[0020] Step 1: obtaining the position information of the brick surface to be laid in the coke oven masonry;

[0021] Step 2: Adjust the position of the robot according to the brick surface position information so that the robot's operating range at least covers the area of ​​the brick surface to be laid;

[0022] Step 3: Obtain visual information of the robot, the discharge port, and the coke oven masonry;

[0023] Step 4: Plan the robot's motion trajectory based on the visual information, calculate the robot's angular error and position error based on the expected posture of the discharge port, and compensate the robot's joint angle and position based on the angular error and position error, so that the robot drives the discharge port to accurately lay the specified brick surface according to the predetermined motion trajectory, and at the same time control the discharge port to evenly extrude the mud according to the preset flow rate and pressure.

[0024] As further preferred, step three includes the following steps:

[0025] (31) planning the robot's motion trajectory based on the visual information;

[0026] (32) The robot performs motion according to the motion trajectory;

[0027] (33) The visual module obtains the actual point cloud data of the robot clamping the discharge port at the current time, and calculates the robot's angle error and position error based on the expected posture;

[0028] (34) Compensating the robot's joint angle and position according to the angle error and the feature point position error so that the robot moves to the desired position by holding the discharge port;

[0029] (35) The mud laying module controls the discharge port to uniformly extrude mud according to the preset flow rate and pressure, so as to accurately lay the mud on the specified brick surface.

[0030] As a further preferred embodiment, the angle error calculation model includes:

[0031] Δθ i =R·mod(R·sign(P di ×P ci )α i ,2π),i=1,...,n

[0032] Where n is the number of angular features, α i =arccos((P ci ·P di ) / (|P ci ||P di |)) is the angle value of the straight line feature in its respective polar coordinate system, P di is the expected feature point position coordinate, P ci is the coordinate of the current feature point, sign(P di×P ci ) is the angle error direction, R is the direction control parameter;

[0033] The position error includes:

[0034] e xi (t) = x ci (t)-x di ,i=1,...,n

[0035] e yi (t) = y ci (t)-y di ,i=1,...,n

[0036] In the formula, (x ci (t),y ci (t)) and (x di ,y di ), i=1,...,n are the current position and expected position of the feature point respectively, e xi (t), e yi (t) are the error of the feature point in the x-axis direction and the y-axis direction in the image coordinate system.

[0037] In general, the above technical solutions conceived by the present invention have the following technical advantages compared with the existing technology:

[0038] 1. The present invention constructs a robot-based brick surface targeted mud precision laying system, which integrates multiple subsystems such as a robot, a mud conveying unit and a mud laying unit, a walking and lifting mechanism, a visual system, and a control system. Through the organic combination of each subsystem, an efficient, flexible and precise overall architecture is formed. The robot is responsible for performing the mud laying operation on the brick surface. Its lightweight structure and high-precision positioning capability enable it to achieve precise mud laying on the side and top surfaces of bricks in a complex masonry environment without causing excessive interference to the already built structure. The mud conveying unit and the mud laying unit ensure the continuous, stable supply and precise extrusion of the mud. The walking and lifting mechanism expands the operating range and flexibility of the robot. The visual system provides accurate brick surface information, and the control system is responsible for coordinating the work of each component. This overall architecture enables the system to independently complete the task of laying mud on brick surfaces without relying on traditional large robots or other complex integrated masonry equipment, improving the system's adaptability and pertinence, and solving problems such as inflexible operation and limited precision caused by the high integration of mud laying functions with other functions in traditional masonry systems. It significantly improves the efficiency and quality of mud laying throughout the entire masonry process, while reducing the requirements for the overall construction environment, providing a more targeted and efficient solution for various types of masonry structure masonry projects.

[0039] 2. The present invention is equipped with a three-dimensional vision system, and suitable visual sensors, such as binocular vision cameras, monocular vision cameras, or depth cameras, can be selected according to actual needs. These visual sensors can accurately obtain three-dimensional geometric information of the side and top surfaces of bricks, including contour features, position, and shape details. Regardless of which visual sensor is selected, the image processing unit will fuse and extract the information it collects and transmit it to the control system. Based on this information, the control system uses a built-in optimization algorithm to calculate the optimal movement distance of the brick surface in each direction, angle adjustment information, and the precise amount and speed of mud extrusion required for the corresponding position. Using this technology, the robot can accurately target mud laying on brick surfaces of different shapes, sizes, and positions, solving the problems of uneven mud distribution and the inability to accurately adjust the laying amount according to the characteristics of the brick surface in traditional mud laying. It achieves accurate and uniform mud laying on the brick surface, ensures consistent mud layer thickness, and significantly improves the quality of masonry.

[0040] 3. The present invention is provided with flow sensors and pressure sensors in the mud conveying unit and the mud laying unit, which are distributed in key positions such as the fluid diverter, feed pipe and discharge port. Based on the real-time feedback of these sensors, the control system uses a closed-loop control algorithm to dynamically adjust the motor speed and the thrust of the cylinder or electric cylinder to achieve precise control of the mud flow and pressure. This ensures that the mud is extruded on the brick surface with a uniform and stable flow rate and pressure, and the accuracy of the mud extrusion amount is controlled at the milliliter / second level to form a high-quality mud layer. At the same time, according to the different conditions of the brick surface (such as roughness, water absorption, etc.), the pressure and flow can be adjusted in real time to adapt the mud extrusion process to different brick surface conditions, meet the diverse masonry process requirements, and avoid masonry quality problems caused by poor control of mud flow and pressure.

[0041] 4. This invention utilizes a horizontally movable travel axis and a vertical column that can be raised and lowered. This travel and lifting mechanism enables flexible horizontal and vertical movement of the robot. Furthermore, the rotation function at the bottom of the vertical column enables the robot to operate from multiple angles. This design allows the robot to easily reach different positions and heights within the masonry area, approaching the sides and tops of bricks from a variety of angles. This expands the robot's operational range and flexibility, resolving the spatial and angular constraints of robot operation in complex masonry environments. This ensures comprehensive coverage of the brick surface throughout the entire masonry area, improving construction efficiency and system adaptability.

[0042] 5. The present invention's slurry delivery system, consisting of a hopper, motor, spiral blades, and fluid diverter, uses a motor-driven spiral blade to continuously deliver slurry from the hopper to the feed pipe, providing a continuous and stable supply for slurry laying. This continuous supply mechanism, combined with the unique design of the fluid diverter, ensures smooth slurry delivery, avoiding the potential interruptions, blockages, and uneven supply issues that can occur in traditional slurry supply systems. Furthermore, the motor speed can be adjusted according to construction progress and requirements, flexibly controlling the slurry supply rate to meet the requirements of slurry laying at different construction speeds and improving the system's overall efficiency.

[0043] 6. The entire system of the present invention is coordinated and managed in a unified manner through a control system. The control system receives information from the visual system, sensors, and various components, and performs data processing and logical judgment. According to different brick surface information and construction requirements, the robot's movements, the working status of the mud conveying unit and the mud laying unit, and the operation of the walking and lifting mechanisms are precisely controlled. Through efficient information interaction and collaborative work between the various components, an automated operation process from brick surface information collection, robot positioning, mud extrusion to reset is realized, forming a highly intelligent operation closed loop. This collaborative and intelligent control method greatly reduces human intervention, improves the system's degree of automation and operating efficiency, avoids the influence of human factors on construction quality, and ensures the stable operation of the entire system and high-quality mud laying operations.

[0044] 7. The present invention integrates an obstacle avoidance system, which uses a three-dimensional vision system or additional obstacle avoidance sensors (such as ultrasonic sensors, lidar, etc.) to monitor the robot's surrounding environment in real time. Through precise environmental perception and path planning, it avoids the robot from colliding with other objects in a complex masonry environment, ensuring the safe operation of the robot and the continuity of the construction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 1 is a schematic elevation view of a robot-based brick surface targeted slurry precision laying system according to the present invention;

[0046] Figure 2 for Figure 1 3D schematic diagram of

[0047] Figure 3 It is a schematic diagram of the robot-based brick surface targeted mud precision laying system of the present invention laying mud on the side of the coke oven masonry.

[0048] In all the drawings, the same figure marks represent the same technical features, specifically: 1-liftable bracket, 2-visual system, 3-robot, 4-lifting mechanism, 5-power drive mechanism, 6-walking mechanism, 7-first fluid diverter, 8-discharge port, 9-feed pipe, 10-mud conveying unit, 11-hopper, 12-motor, 13-mud laying unit, 14-spiral blade, 16-coke oven masonry, 17-second fluid diverter. DETAILED DESCRIPTION

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0050] like Figures 1 to 3 As shown, an embodiment of the present invention provides a robot-based brick surface targeted mud precision laying system, including: a robot 3; a mud laying module 13, the mud laying module 13 including a feeding unit and a discharge port 8 connected to the feeding unit, the feeding unit being used to deliver mud of specified pressure to the discharge port 8, the discharge port 8 being clamped at the end of the robot 3; a visual module 2, being used to obtain visual information of the robot 3, the discharge port 8 and the coke oven masonry 16; a control module, being used to plan the motion trajectory of the robot 3 according to the visual information, calculate the angular error and position error of the robot 3 according to the expected posture of the discharge port 8, and compensate the joint angle and position of the robot 3 according to the angular error and position error, so that the robot 3 drives the discharge port 8 to precisely lay the specified brick surface according to the predetermined motion trajectory.

[0051] Based on any of the above embodiments or a combination of multiple embodiments, in this embodiment, the feeding unit includes a hopper 11 for placing mud, and a first fluid diverter 17 is provided at the bottom of the hopper 11. The first fluid diverter 17 is connected to the discharge port 8 through a feed pipe 9, and is used to divert the mud in the hopper 11 to the feed pipe 9. A second fluid diverter 7 is also provided between the feed pipe 9 and the discharge port 8.

[0052] Based on any of the above embodiments or a combination of multiple embodiments, in this embodiment, a spiral blade 14 is provided between the first fluid diverter 17 and the second fluid diverter 7. The spiral blade 14 is a shaftless spiral blade, which is arranged along the central axis of the feed pipe 9 and is driven to rotate by a motor.

[0053] Based on any of the above embodiments or a combination of multiple embodiments, in this embodiment, a gradual buffer chamber is provided in the middle of the feed pipe 9, and the inner diameter of the gradual buffer chamber gradually decreases along the direction from the first fluid diverter 17 to the second fluid diverter 7 to adjust the flow rate and pressure of the mud and discharge the gas in the mud at the same time.

[0054] Based on any of the above embodiments or a combination of multiple embodiments, in this embodiment, a pre-pressure limiting valve is provided between the first fluid diverter 17 and the spiral blade 14 , and the pre-pressure limiting valve is used to open when the mud pressure is greater than a preset value to transport mud to the spiral blade 14 .

[0055] Based on any of the above embodiments or a combination of multiple embodiments, this embodiment also includes a walking drive module for driving the robot 3 to walk, and the walking drive module includes a horizontal walking unit 6 and a lifting unit 4. The horizontal walking unit 6 is used to drive the robot 3 to move horizontally along the coke oven masonry 16, and the lifting unit 4 is used to drive the robot 3 to move vertically along the coke oven masonry 16.

[0056] Based on any of the above embodiments or a combination of multiple embodiments, in this embodiment, the horizontal walking unit 6 includes a horizontal support member and a horizontal screw drive assembly provided on the horizontal support member, and a support plate is provided on the sliding nut of the horizontal screw drive assembly; the lifting unit 4 is provided on the support plate, and the lifting unit 4 includes a vertical support member and a vertical screw drive assembly provided on the vertical support member, and the sliding nut of the vertical screw drive assembly is fixedly connected to the robot 3.

[0057] Based on any of the above embodiments or a combination of multiple embodiments, in this embodiment, the control module is used to perform the following actions:

[0058] The position of the robot 3 is adjusted according to the brick surface position information, so that the operating range of the robot 3 at least covers the range of the brick surface to be laid; the visual information of the robot 3, the discharge port 8 and the coke oven masonry 16 is obtained; the motion trajectory of the robot 3 is planned according to the visual information, the angular error and position error of the robot 3 are calculated according to the expected posture of the discharge port 8, and the joint angle and position of the robot 3 are compensated according to the angular error and position error, so that the robot 3 drives the discharge port 8 to accurately lay the specified brick surface according to the predetermined motion trajectory, and at the same time controls the discharge port 8 to evenly extrude the mud according to the preset flow rate and pressure.

[0059] More specifically, the system comprises several key components: a robot, a slurry delivery unit, a slurry application unit, a travel mechanism, a lifting mechanism, a vision system, and a control system. These components work closely together to precisely and evenly apply slurry to both the side and top surfaces of bricks (including the sides where bricks are fixed during the laying process and the top surfaces where bricks will be subsequently laid).

[0060] Robot: This lightweight robot offers high flexibility and high-precision positioning capabilities. Its lightweight structure makes it suitable for operation in relatively confined or complex masonry environments. A support is mounted on the end of the robot, onto which a pneumatic or electric cylinder is attached. This cylinder or electric cylinder provides the pressure required to push the slurry. The robot is equipped with an advanced 3D vision system that accurately identifies the position, shape, size, and other surface features of the sides and tops of bricks, providing precise target information for subsequent slurry placement operations. The robot is also equipped with an obstacle avoidance system. This system uses environmental information provided by the 3D vision system or additional obstacle avoidance sensors (such as ultrasonic sensors and lidar) to monitor its surroundings in real time during operation. By identifying and analyzing the position and shape of surrounding objects, the system plans a safe movement path, ensuring that the robot avoids collisions with other objects during movement and operation. When a potential collision risk is detected, the obstacle avoidance system sends a signal to the control system, which adjusts the robot's trajectory to avoid a collision.

[0061] Mud conveying unit: It mainly consists of a hopper, a motor, a spiral blade and a fluid diverter. The hopper is located in the construction area and is used to store mud raw materials. The motor is connected to the fluid diverter at the bottom of the hopper. When the motor is working, it drives the spiral blade to rotate and transports the mud from the hopper to the feed pipe. Specifically, the feed pipe is connected to one end of the fluid diverter at the bottom of the hopper. A spiral blade for spiral conveying is provided inside the fluid diverter. The motor drives the spiral blade to rotate, thereby realizing the process of conveying the mud from the hopper to the feed pipe. The fluid diverter plays the role of distributing the mud. One end of the fluid diverter is connected to the outlet of the feed pipe, and the other end is connected to the cylinder body of the air cylinder or electric cylinder. Flow sensors are provided at different ports for real-time monitoring of the flow of mud in each part. Based on the data feedback from the flow sensor, the control system can accurately control the amount of mud delivered. This unit provides a continuous and stable mud supply for subsequent mud laying. At the same time, through real-time monitoring of the flow sensor and feedback adjustment of the control system, it avoids the problems of flow interruption, blockage or uneven supply that may occur in traditional mud supply systems. It can also adjust the motor speed according to the construction progress and needs, and flexibly control the mud supply speed to meet the mud laying needs at different construction speeds, thereby improving the overall work efficiency of the system.

[0062] The slurry application unit primarily consists of a second fluid diverter, a feed pipe, a discharge port, and a pneumatic or electric cylinder. One port of the second fluid diverter is connected to the outlet of the feed pipe, and the other port is connected to the flat-mouth discharge port. The cylinder or electric cylinder's body is connected to the second fluid diverter, and its push rod is located within the channel of the second fluid diverter. The feed pipe receives slurry from the slurry delivery unit. The push rod of the pneumatic or electric cylinder retracts and extends, pushing the slurry in the feed pipe to the flat-mouth discharge port, thereby extruding the slurry. Pressure sensors are installed near the feed pipe and discharge port to measure the slurry pressure at different locations to prevent problems such as blockage or pressure anomalies. Based on the pressure sensor data, the control system dynamically adjusts the thrust of the pneumatic or electric cylinder to ensure stable and uniform slurry extrusion. This allows precise control of the slurry extrusion pressure based on varying brick surface conditions (such as roughness and water absorption), ensuring uniform and stable slurry extrusion across the brick surface, resulting in a consistent, high-quality slurry layer.

[0063] Travel Mechanism: A horizontally movable travel axis connected to the robot enables free horizontal movement, expanding its operating range to accommodate brick laying requirements in various locations. This allows the robot to flexibly move horizontally, facilitating positioning in different masonry areas. This provides horizontal position adjustment for mortar laying operations, ensuring the system can cover a wider range of brick surfaces, improving construction efficiency and system adaptability.

[0064] Lifting mechanism: A vertical column is mounted at the bottom of the robot, allowing for lifting and movement. A rotating mechanism is also located at the bottom. This mechanism allows the robot to approach the brick surface from different heights. Combined with the walking and rotating mechanisms, the robot can approach the sides and tops of bricks from different heights and angles, ensuring comprehensive mortar application on both sides and tops. This addresses the issue of robot operation being limited by space and angles in complex masonry environments, ensuring complete coverage of the brick surface throughout the entire masonry area.

[0065] Vision System: Composed of an adjustable mounting bracket, multiple visual sensors, and an image processing unit, it is designed to provide accurate and comprehensive visual information about the brick surface for the entire robot-based targeted mortar placement system. This enables effective visual guidance of the robot and mortar placement device, ensuring precise operation at every stage. This vision system is independent of any other possible visual navigation systems of the masonry robot and is specifically designed for the precise operation of this system, such as brick surface search and mortar placement position alignment. It works in conjunction with the visual navigation systems of other masonry robots, which control the robot's position from its origin to the masonry position, to enhance the system's intelligence and operational accuracy.

[0066] Multiple types of visual sensors: This system can flexibly select different types of visual sensors according to actual needs. These sensors each have unique advantages and can meet the requirements of obtaining visual information of the brick surface in a targeted manner to achieve precise guidance of robots and mud laying equipment.

[0067] For example, a binocular vision camera can be used. Based on the principle of triangulation, two cameras are used to simultaneously capture the brick surface from different angles to obtain a pair of images with parallax. Subsequently, professional image matching and depth calculation algorithms are used to accurately restore the three-dimensional geometric shape of the brick surface and clearly extract its outer contour information, including the detailed geometric features of the contour, the angular relationship between the surfaces, and the positions of key feature points. These data provide key basic data for subsequent spatial positioning and posture analysis, helping to accurately determine key parameters such as the displacement and rotation angle of the brick surface in three-dimensional space, thereby guiding the robot to accurately reach the brick surface position and laying the foundation for the precise positioning and adaptability adjustment of the mud discharge port, ensuring that the discharge port and the mud laying position on the brick surface are precisely aligned.

[0068] A monocular camera is also an option. Relying on its efficient image processing capabilities, combined with a pre-stored library of brick surface models and relevant prior knowledge, it uses feature extraction and shape matching to quickly identify the approximate type, shape, and position of brick surfaces within a two-dimensional plane. In scenarios with relatively moderate precision requirements, such as the initial stages of brick surface search, a monocular camera can serve as an auxiliary tool for preliminary screening and rough positioning of brick surfaces within the work area, improving overall system efficiency, narrowing the target range for subsequent, more precise visual measurement steps, and optimizing the allocation of computing resources and processing flows.

[0069] Depth cameras can also be configured on demand. They can directly obtain depth information of the brick surface and generate depth images or point cloud data, making up for the shortcomings of simple two-dimensional images in reflecting spatial structure. The depth information is integrated with the two-dimensional image data collected by binocular and monocular vision cameras (if configured) to further enhance the system's understanding of the three-dimensional structure of the brick surface, allowing the system to more accurately grasp the concave and convex conditions of the brick surface, the three-dimensional layering of the edges, and the spatial relationship between each part. Especially when dealing with brick surfaces with complex curves or irregular depressions, more detailed data support is provided for accurate contour extraction, equal division of points, and shape analysis, ensuring that the mud laying system can make adaptive adjustments based on the actual conditions of the brick surface and ensure uniform and accurate laying of the mud on the brick surface.

[0070] Image processing unit:

[0071] As the core data processing module of the vision system, the image processing unit is responsible for fusing, processing, analyzing, and extracting key information from the image data collected by various visual sensors. It integrates data from various visual sensors, applying advanced algorithms to remove noise, enhance image clarity, and accurately identify key features of the brick surface. By deeply mining this data, it provides the control system with precise and real-time visual feedback, such as the real-time position, posture, and shape details of the brick surface. This information enables the control system to make accurate decisions and coordinate the smooth operation of various components.

[0072] Control system: Data acquisition and processing: Receives information such as the shape and size of the brick surface from the visual system. It can also collect data from the flow sensors and pressure sensors in the mud conveying unit and the mud laying unit, as well as information such as the working status of the cylinder or electric cylinder. Based on this large amount of data, the built-in optimization algorithm model is used to perform complex mathematical operations and logical judgments. First, by analyzing the size and shape data of the brick surface, the various surfaces of the brick surface that need to be laid with mud, as well as the area, shape characteristics, and other information of each surface are determined; then, based on this information and the preset mud distribution rules, the optimal movement distance of the brick surface in each direction and the precise amount of mud extrusion corresponding to each position are calculated, and detailed control instructions are generated to provide accurate parameter basis for subsequent mud laying operations.

[0073] Equipment Coordination Control: Based on calculated parameters, speed control commands are sent to the motor, thrust control commands are sent to the pneumatic or electric cylinder, and motion commands are sent to the robot, ensuring coordinated operation among all components. Throughout the entire operation, the control system monitors the operating status of each component in real time. If any anomalies are detected, fault diagnosis and emergency response are carried out promptly to ensure smooth construction.

[0074] At the same time, the control system receives information sent by the obstacle avoidance system. When the obstacle avoidance system detects a potential collision risk, it adjusts the robot's motion trajectory and speed according to the risk information to ensure the robot's safe operation.

[0075] According to another aspect of the present invention, a robot-based method for precise grouting on brick surfaces is provided, comprising the following steps:

[0076] Step 1: obtaining the position information of the brick surface to be laid in the coke oven masonry;

[0077] Step 2: Adjust the position of the robot 3 according to the brick surface position information so that the operation range of the robot 3 at least covers the range of the brick surface to be laid;

[0078] Step 3: Obtain visual information of the robot 3, the discharge port 8, and the coke oven masonry 16; specifically, in this step:

[0079] (31) planning the motion trajectory of the robot 3 according to the visual information;

[0080] (32) Robot 3 performs motion according to the motion trajectory;

[0081] (33) The visual module 2 obtains the actual point cloud data of the robot 3 holding the discharge port 8 at the current time, and calculates the angle error and position error of the robot 3 according to the expected posture;

[0082] (34) Compensating the joint angle and position of the robot 3 according to the angle error and the characteristic point position error, so that the robot 3 moves to the desired position by holding the discharge port 8;

[0083] (35) The mud laying module 13 controls the discharge port 8 to uniformly extrude mud according to the preset flow rate and pressure, so as to accurately lay the mud on the specified brick surface.

[0084] Step 4: Plan the motion trajectory of robot 3 based on the visual information, calculate the angular error and position error of robot 3 based on the desired position of discharge port 8, and compensate the joint angle and position of robot 3 based on the angular error and position error, so that robot 3 drives discharge port 8 to accurately lay the specified brick surface according to the predetermined motion trajectory, and at the same time controls discharge port 8 to uniformly extrude mud according to the preset flow rate and pressure. The angular error calculation model includes:

[0085] Δθ i =R·mod(R·sign(P di ×P ci )α i ,2π),i=1,...,n

[0086] Where n is the number of angular features, α i =arccos((P ci ·P di ) / (|P ci ||P di |)) is the angle value of the straight line feature in its respective polar coordinate system, P di is the expected feature point position coordinate, P ci is the coordinate of the current feature point, sign(P di ×P ci ) is the angle error direction, R is the direction control parameter;

[0087] The position error includes:

[0088] e xi (t) = x ci (t)-x di ,i=1,...,n

[0089] e yi (t) = y ci (t)-y di ,i=1,...,n

[0090] In the formula, (x ci (t),y ci (t)) and (x di ,y di ), i=1,...,n are the current position and expected position of the feature point respectively, e xi (t), e yi (t) are the error of the feature point in the x-axis direction and the y-axis direction in the image coordinate system.

[0091] In this step, the actual image of the robot at the current moment is collected, and the angular error and feature point position error of the robot are calculated based on the expected image and the actual image of the robot. Based on the angular error, the straight line feature is used to control the posture of the robot. At the same time, dynamic error mapping is performed based on the feature point position error. According to the mapping relationship, the feature point position change caused by the posture movement is compensated, and then the robot joint speed is updated according to the posture error. More specifically, in this solution, a straight line angle feature is constructed based on the point cloud data of the robot or the discharge port obtained by the visual device, and the angle error value is calculated based on the straight line angle feature in the expected image and the current actual image. More specifically, the four feature points are connected in sequence to form a quadrilateral, and the straight line where the edge of the quadrilateral is located is used as the straight line feature for controlling the posture. More specifically, the four feature points are connected in sequence to form a quadrilateral, and the straight line where the edge of the quadrilateral is located is used as the straight line feature for controlling the posture; the foot of the perpendicular line through the origin P of the straight line is used. i =(x Pi ,y Pi )'s polar coordinates [ρ i ,θ i ] T To parameterize the line feature, where x Pi is the x-direction coordinate of the foot of the perpendicular in the Cartesian coordinate system, y Pi is the y-direction coordinate of the foot of the perpendicular in the Cartesian coordinate system, ρ i P i Distance to the origin, θ i P i The angle between the vector and the x-axis of the image coordinate system; the polar coordinate corresponding to the perpendicular foot (ρ i ,θ i ) is used as the servo variable, and the relationship between the linear angle feature change speed and the camera angular velocity is constructed as follows:

[0092]

[0093] Where, is the angle change rate, ω c is the angular velocity of the camera.

[0094] In the above steps, it is necessary to use a visual device to take pictures of the four point features to obtain a desired image and obtain the coordinates of the point features in the image coordinate system through an image processing algorithm.

[0095] The straight line where the four feature points are connected to form the edge of the quadrilateral is used as the straight line feature to control the posture. The foot of the perpendicular line through the origin is used as the foot of the perpendicular line P. i =(x Pi ,y Pi )'s polar coordinates [ρ i ,θ i ] T To parameterize the straight line feature. According to the geometric properties of the perpendicular line, it can be calculated

[0096]

[0097] The polar coordinates corresponding to the foot of the perpendicular (ρ i ,θ i ) as the servo variable, ρ i P i Distance to the origin, θ i P i The angle between the vector and the x-axis of the image coordinate system is calculated as follows:

[0098]

[0099] When the camera optical axis is approximately perpendicular to the plane where the straight line feature is located, the relationship between the camera angular velocity and the angular change rate of the straight line feature is:

[0100]

[0101] θ i This is the angle characteristic of the straight line. i When expressing the polar coordinates of the characteristic angle of a straight line, the respective polar axes are established by connecting the foot of the perpendicular to the imaging plane at the desired position.

[0102] Based on the above scheme, after obtaining the angle error and position error, a forward kinematic model of the robot is established to convert the joint angles into the position and posture of the end effector. The Denavit-Hartenberg (DH) parameter method is used to define the parameters of each joint and link of the robot. The position and posture of the end effector are calculated using the rotation transformation matrix and the translation transformation matrix. According to the Newton-Euler equation, an inverse dynamics model of the robot is established to calculate the torque required to drive the joint motor. Considering factors such as the robot's inertia, centrifugal force, Coriolis force, and gravity, the formula is as follows:

[0103]

[0104] Among them, τ is the joint torque vector, M(q) is the inertia matrix, which is related to the joint angle vector q. are the centrifugal force and Coriolis force terms, and the joint angle vector q and joint velocity vector G(q) is the gravity term, which is related to the joint angle vector q.

[0105] Based on the above scheme, in this embodiment, a real-time data interaction channel is established between the robot posture control system and the mud adjustment system, so that the two subsystems can share information, such as the position and angle deviation of the robot end, the flow rate and pressure deviation of the mud, etc. The adaptive dynamic programming (ADP) control algorithm is used to integrate the control of the robot posture and mud parameters. The ADP algorithm can adaptively adjust the control strategy by learning the system dynamics online to cope with the uncertainty in complex environments. A unified controller is constructed using a neural network, the input of which includes posture error and mud parameter error, and the output is the drive signal of the joint motor and the control signal of the mud adjustment device. The neural network optimizes the control effect by continuously learning and adjusting the weights. A unified optimization control model is established, with the robot posture and mud parameters as control variables, and the flatness, density and bonding strength of the bricks as the objective function. The optimal control strategy is obtained by solving the optimization problem.

[0106] The control strategy is as follows:

[0107] Hierarchical control strategy: Implement a hierarchical control strategy, with the top layer responsible for overall planning and coordination, the middle layer responsible for the integrated control of posture control and mud regulation, and the bottom layer responsible for the specific actuator drive.

[0108] Real-time feedback adjustment: During the control process, the feedback information of posture and mud parameters is monitored in real time, and the control strategy is adjusted in time according to the deviation to ensure the stability and adaptability of the system.

[0109] Specifically, an evaluation network and a strategy network are designed to evaluate the quality of the current control strategy and generate a control strategy, respectively. The network is continuously trained through real-time data, and the network weights and biases are updated to adapt to changes in the environment. The input layer of the neural network is defined as the posture error and the mud parameter error, and the output layer is defined as the drive signal of the joint motor and the control signal of the mud regulating device. The neural network is trained using historical data and real-time data, and an optimization algorithm (such as the gradient descent method) is used to minimize the error. The objective function is clarified, such as the flatness and density of the bricks and the bonding strength of the mud, and these indicators are quantified into computable mathematical expressions. The constraints of the control variables are set, such as the limits of joint angles and speeds, the range of mud flow and pressure, etc. Optimization algorithms (such as particle swarm optimization algorithms, genetic algorithms, etc.) are used to solve the optimization problem and obtain the optimal control strategy.

[0110] More specifically, based on any of the above systems or methods, the workflow of this system is as follows:

[0111] First, the slurry is added to the hopper. The motor is then started, driving the spiral blades in the fluid diverter at the bottom of the hopper, continuously feeding the slurry into the feed pipe. The robot uses a 3D vision system to scan the masonry area, accurately locating the sides and tops of the bricks where slurry is to be applied, including those already laid or awaiting placement. The vision system also transmits this brick surface information to the control system, which analyzes and processes it to determine the brick's position, shape, dimensions, and other characteristics, calculating the parameters required for slurry application. The robot then moves to the target brick surface using the travel axis and vertical columns, based on the visual information and instructions from the control system. During this process, the obstacle avoidance system continuously monitors the surrounding environment. If a potential collision risk is detected, it sends a signal to the control system, which adjusts the robot's trajectory. Once the robot reaches its target position, a pneumatic or electric cylinder activates, pushing the slurry from the feed pipe out through the discharge port, evenly applying the slurry to the brick surface. During the entire process, flow sensors and pressure sensors monitor the flow and pressure of the mud in real time and feed the data back to the control system. Based on the feedback information, the control system dynamically adjusts the motor speed, the thrust of the cylinder or electric cylinder, and the position and motion trajectory of the robot to ensure targeted and precise mud laying on the brick surface.

[0112] Robot-based precise grouting method for brick surfaces:

[0113] (1) System startup and initialization

[0114] At system startup, each component automatically initiates a self-test. The robot's arm extends and its joints move to perform initialization operations, ensuring that their range of motion, accuracy, and flexibility meet preset requirements. Key actuator components, such as the robot's drive motor and pneumatic or electric cylinders, undergo performance tests, including checking the motor's speed and torque, as well as the cylinder's retractable performance.

[0115] At the same time, the robot's obstacle avoidance system starts a self-test program to ensure that its obstacle avoidance sensors (such as ultrasonic sensors, lidar, etc.) are functioning normally and can accurately perceive the surrounding environment information, and check the effectiveness of the obstacle avoidance algorithm to ensure that the obstacle avoidance operation can be performed normally in subsequent operations.

[0116] The horizontal travel axis and vertical column of the walking and lifting mechanism are initialized to test the smoothness and accuracy of their horizontal and lifting movements. The rotation angle range of the rotating mechanism is calibrated to ensure that the predetermined horizontal and vertical movement range and rotation angle range can be achieved.

[0117] Check the valves and connections of the hoppers in the slurry conveying and slurry laying units for tightness. Perform a no-load test on the motor to check for smooth rotation and normal speed. Check the feed pipe and discharge port for blockage and the fluid diverter for tight connections. Also, check the installation and rotational flexibility of the spiral blades in the feed pipe.

[0118] The vision system starts, calibrates the lens, initializes the image acquisition and processing program, and the image processing unit loads the image processing algorithm and image database to prepare for the subsequent brick surface information collection and analysis.

[0119] The control system loads preset algorithm models and parameter settings, including robot motion control algorithms, mud flow and pressure control algorithms, obstacle avoidance control algorithms, etc., starts data communication functions with other components, prepares to receive massive data from the vision system and sensors, and starts subsequent control and calculation operations to ensure that the entire system reaches the best operating state in the initial state.

[0120] (2) Brick surface information collection and transmission

[0121] The robot's 3D vision system is activated to scan the sides and tops of the bricks within the masonry area. Multiple visual sensors, such as binocular cameras, monocular cameras, or depth cameras (depending on the system configuration), collect image information of the brick surfaces from different angles.

[0122] Based on the principle of triangulation, the binocular vision camera simultaneously captures the brick surface from different angles, acquiring pairs of images with parallax. Then, using image matching and depth calculation algorithms, it accurately restores the three-dimensional geometry of the brick surface and extracts its outer contour information, including its detailed geometric features, the angular relationships between the surfaces, and the locations of key feature points. The monocular vision camera, relying on its image processing capabilities, combined with a pre-stored brick surface model library and prior knowledge, rapidly identifies and roughly locates the brick surface, determining its approximate type, shape, and position within the two-dimensional plane. The depth camera directly captures the depth information of the brick surface, generating depth images or point cloud data to compensate for the shortcomings of two-dimensional images in representing spatial structure and provide the system with richer brick surface information.

[0123] The image processing unit fuses data from different visual sensors to remove noise, enhance image clarity, and accurately extract key features of the brick surface, such as position, posture, shape details, etc., and transmits this brick surface information to the control system in real time, providing an accurate basis for subsequent operations.

[0124] (3) Data processing and decision-making calculation

[0125] Upon receiving the brick surface information, the control system immediately activates a built-in optimization algorithm. This algorithm comprehensively considers the pre-stored correspondence between brick surface information and mortar laying parameters for different brick types and masonry scenarios (based on extensive experimental and empirical data), while also incorporating current masonry process requirements (such as mortar thickness and brick surface flatness requirements) and mortar characteristic parameters (such as mortar viscosity, fluidity, and setting time).

[0126] Based on the shape and size of the brick surface and the preset slurry placement rules, the robot calculates the optimal movement distance and angle adjustment information for each brick surface in each direction, as well as the precise slurry extrusion volume and speed required at the corresponding position. Based on these calculation results, detailed control instructions are generated, including the robot's motion instructions, and control instructions for the slurry delivery and placement units, providing accurate parameters for subsequent slurry placement operations.

[0127] During this process, obstacle avoidance factors are taken into account, environmental information and the location of potential obstacles are incorporated into the calculation, and a safe movement path is planned for the robot to ensure that the robot avoids collision risks when performing subsequent operations.

[0128] (4) Robot position adjustment and positioning

[0129] The control system sends precise control signals to the robot based on previously calculated motion parameters. Upon receiving these signals, the robot precisely adjusts its position through its travel and lifting mechanisms, aligning the slurry outlet at its end with the side or top surface of the brick.

[0130] During the movement of the robot, its position sensor feeds back the current position information to the control system in real time. The control system makes real-time adjustments and corrections based on this feedback information to ensure that the positioning accuracy between the robot and the brick surface reaches the millimeter level.

[0131] At the same time, the obstacle avoidance system continuously monitors the robot's surroundings and transmits environmental information to the control system. The control system adjusts the robot's motion path in real time based on the environmental information and the robot's current position and motion trajectory, ensuring that the robot avoids obstacles during movement and reaches the target position safely, making full preparations for subsequent mud extrusion operations.

[0132] In this step, the robot's motion trajectory is planned based on the visual information, the robot's angular error and position error are calculated based on the desired posture of the discharge port, and the robot's joint angle and position are compensated based on the angular error and position error, so that the robot drives the discharge port to accurately lay the specified brick surface according to the predetermined motion trajectory, and at the same time controls the discharge port to uniformly extrude mud according to the preset flow rate and pressure. The robot's motion trajectory is planned based on the visual information; the robot performs motion actions according to the motion trajectory; the visual module obtains the actual point cloud data of the robot clamping the discharge port at the current time, and calculates the robot's angular error and position error based on the desired posture; the robot's joint angle and position are compensated based on the angular error and feature point position error, so that the robot clamps the discharge port and moves to the desired posture; the mud laying module controls the discharge port to uniformly extrude mud according to the preset flow rate and pressure, so as to accurately lay the specified brick surface.

[0133] (5) Mud extrusion and flow control

[0134] Once the robot is precisely positioned, the control system sends a start command to the motors and pneumatic or electric cylinders in the slurry delivery and placement units, initiating the slurry extrusion process. The motor drives the spiral blades, transferring the slurry from the hopper to the feed pipe. The pneumatic or electric cylinder then pushes the slurry out of the feed pipe toward the discharge port.

[0135] At the same time, the control system precisely controls the slurry flow and pressure based on the calculated slurry extrusion volume and speed by adjusting the motor speed and the thrust of the pneumatic or electric cylinder. Flow sensors and pressure sensors installed on the fluid diverter, feed pipe, and discharge port monitor the slurry flow and pressure in real time. The control system utilizes a closed-loop control algorithm to dynamically adjust the motor speed and the thrust of the pneumatic or electric cylinder based on this feedback information, ensuring that the slurry is extruded onto the brick surface at a uniform and stable flow and pressure, forming a slurry layer of consistent thickness and reliable quality.

[0136] During the mud extrusion process, the obstacle avoidance system continuously monitors the robot's surroundings. When it detects a possible obstacle approaching, it feeds the information back to the control system. The control system may suspend the mud extrusion operation, adjust the robot's position, or wait for the obstacle to be removed to ensure safe operation.

[0137] (6) Completion and Reset

[0138] When the slurry laying on the brick surface reaches a predetermined area or the extrusion amount meets preset completion conditions such as the calculated value, the control system sends a stop command to the slurry conveying unit and the slurry laying unit, the motor stops rotating, and the cylinder or electric cylinder stops working.

[0139] Subsequently, the control system sends a reset command to the robot, and the robot moves itself to the initial position or the next brick surface position where mud is to be laid, and at the same time adjusts the mud outlet at its end to the appropriate position to prepare for the next mud laying operation.

[0140] During the robot resetting process, the robot's position sensor feeds back position information to the control system in real time. The control system monitors and adjusts based on this information to ensure accurate resetting of the robot.

[0141] At the same time, the obstacle avoidance system continuously monitors the robot's surroundings to ensure the robot's safe movement during the reset process, avoids collisions with other objects, and ensures the efficient and continuous operation of the entire system.

[0142] (7) System maintenance and cleaning

[0143] Maintain the system regularly or after completing a certain amount of mud laying tasks.

[0144] For the mud conveying unit and mud laying unit, the hopper can be cleaned as needed to prevent the mud from settling and drying out, which will affect the quality and conveying performance of the subsequent mud.

[0145] For the feed pipe and discharge port, special cleaning tools or cleaning fluids can be used to remove residual mud to ensure the smooth flow of mud conveying channels. For example, high-pressure water flushing or special chemical solvents can be used to dissolve residual mud.

[0146] Clean the lenses and sensors of the visual system to ensure that they can collect image information clearly and accurately, and avoid factors such as dust and mud splashing that affect the visual effect.

[0147] Check and maintain the control system's software and hardware, update algorithm models and parameter settings to ensure its performance and stability.

[0148] Clean and calibrate the obstacle avoidance system's sensors to ensure they can accurately perceive their surroundings and avoid obstacles like dust or mud. Also, review and update the obstacle avoidance algorithm to ensure its effectiveness in different environments and scenarios.

[0149] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements 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 robot-based brick surface targeted grouting system, characterized in that: include: Robot (3); A slurry laying module (13), comprising a feeding unit and a discharge port (8) connected to the feeding unit, wherein the feeding unit is used to deliver slurry of a specified pressure to the discharge port (8), and the discharge port (8) is clamped at the end of the robot (3); A visual module (2) for acquiring visual information of the robot (3), the discharge port (8) and the coke oven masonry (16); A control module is used to plan the motion trajectory of the robot (3) according to the visual information, calculate the angle error and position error of the robot (3) according to the expected posture of the discharge port (8), and compensate the joint angle and position of the robot (3) according to the angle error and position error, so that the robot (3) drives the discharge port (8) to accurately lay the specified brick surface according to the predetermined motion trajectory.

2. A robot-based brick surface targeted grouting system according to claim 1, characterized in that: The feeding unit comprises a hopper (11) for placing mud, a first fluid diverter (17) is provided at the bottom of the hopper (11), the first fluid diverter (17) is connected to the discharge port (8) through a feeding pipe (9), and is used to divert the mud in the hopper (11) to the feeding pipe (9), and a second fluid diverter (7) is further provided between the feeding pipe (9) and the discharge port (8).

3. A robot-based brick surface targeted grouting system according to claim 2, characterized in that: A spiral blade (14) is provided between the first fluid diverter (17) and the second fluid diverter (7). The spiral blade (14) is a shaftless spiral blade, which is arranged along the central axis of the feed pipe (9) and is driven to rotate by a motor.

4. A robot-based brick surface targeted grouting system according to claim 3, characterized in that: A gradual buffer chamber is provided in the middle of the feed pipe (9), and the inner diameter of the gradual buffer chamber gradually decreases along the direction from the first fluid diverter (17) to the second fluid diverter (7) to adjust the flow rate and pressure of the mud and discharge the gas in the mud at the same time.

5. A robot-based brick surface targeted grouting system according to claim 4, characterized in that: A pre-pressure limiting valve is provided between the first fluid diverter (17) and the spiral blade (14). The pre-pressure limiting valve is used to open when the mud pressure is greater than a preset value to transport mud to the spiral blade (14).

6. A robot-based brick surface targeted grouting system according to any one of claims 1 to 5, characterized in that: It also includes a walking drive module for driving the robot (3) to walk, the walking drive module including a horizontal walking unit (6) and a lifting unit (4), the horizontal walking unit (6) is used to drive the robot (3) to move in the horizontal direction along the coke oven masonry (16), and the lifting unit (4) is used to drive the robot (3) to move in the vertical direction along the coke oven masonry (16).

7. A robot-based brick surface targeted grouting system according to any one of claims 1 to 6, characterized in that: The horizontal walking unit (6) includes a horizontal support member and a horizontal screw drive assembly provided on the horizontal support member, wherein a support plate is provided on the sliding nut of the horizontal screw drive assembly; the lifting unit (4) is provided on the support plate, wherein the lifting unit (4) includes a vertical support member and a vertical screw drive assembly provided on the vertical support member, wherein the sliding nut of the vertical screw drive assembly is fixedly connected to the robot (3).

8. A robot-based targeted grouting method for brick surfaces, characterized in that: The following steps are involved: Step 1: obtaining the position information of the brick surface to be laid in the coke oven masonry; Step 2: adjusting the position of the robot (3) according to the brick surface position information so that the operation range of the robot (3) at least covers the range of the brick surface to be laid; Step 3, obtaining visual information of the robot (3), the discharge port (8) and the coke oven masonry (16); Step 4: planning the motion trajectory of the robot (3) based on the visual information, calculating the angle error and position error of the robot (3) based on the expected posture of the discharge port (8), and compensating the joint angle and position of the robot (3) based on the angle error and position error, so that the robot (3) drives the discharge port (8) to accurately lay the specified brick surface according to the predetermined motion trajectory, and at the same time controls the discharge port (8) to uniformly extrude the mud according to the preset flow rate and pressure.

9. A robot-based targeted grouting method for brick surfaces according to claim 8, characterized in that: Step three includes the following steps: (31) planning the motion trajectory of the robot (3) according to the visual information; (32) The robot (3) performs motion actions according to the motion trajectory; (33) The vision module (2) obtains the actual point cloud data of the robot (3) holding the discharge port (8) at the current time, and calculates the angle error and position error of the robot (3) according to the expected posture; (34) compensating the joint angle and position of the robot (3) according to the angle error and the characteristic point position error, so that the robot (3) moves to a desired position by clamping the discharge port (8); (35) The mud laying module (13) controls the discharge port (8) to uniformly extrude mud according to the preset flow rate and pressure, so as to accurately lay the mud on the designated brick surface.

10. A robot-based targeted grouting method for brick surfaces according to claim 9, characterized in that: The angle error calculation model includes: Dth i =R·mod(R·sign(P di ×P ci )a i ,2π),i=1,...,n Where n is the number of angle features, α i =arccos((P ci ·P di ) / (|P ci ||P di |)) is the angle value of the straight line feature in its respective polar coordinate system, P di is the expected feature point position coordinate, P ci is the coordinate of the current feature point, sign(P di ×P ci ) is the angle error direction, R is the direction control parameter; The position error includes: e xi (t)=x ci (t)-x di ,i=1,...,n e yi (t)=y ci (t)-y di ,i=1,...,n In the formula, (x ci (t),y ci (t)) and (x di ,y di ), i=1,...,n are the current position and expected position of the feature point respectively, e xi (t), e yi (t) are the error of the feature point in the x-axis direction and the y-axis direction in the image coordinate system.

Images