Unmanned aerial vehicle inkjet system and method and readable storage medium

Abstract

The invention relates to an unmanned aerial vehicle inkjet system and method and a readable storage medium, and the unmanned aerial vehicle inkjet system comprises a sensing positioning module and an inkjet device which are installed on an unmanned aerial vehicle, and a path planning module and a control module which are arranged at a ground end. The control module controls the flight position and the flight attitude of the unmanned aerial vehicle based on the inkjet path, and controls the relative attitude of a nozzle of the inkjet device and the wall surface to be stable so as to control the inkjet device to perform uniform inkjet on the wall surface; and the control module is also used for receiving the second image data, determining a quality defect area based on the second image data, controlling the unmanned aerial vehicle to fly to the quality defect area, and controlling the spray painting device to perform supplementary spray painting on the quality defect area to integrally improve the spray painting quality and efficiency.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a UAV inkjet printing system, method, and readable storage medium. Background Technology

[0002] In the field of exterior wall painting in the construction industry, the current mainstream operation method still heavily relies on manual labor, which poses significant safety risks, health hazards, and efficiency bottlenecks. Exterior wall painting often needs to be completed at heights. Traditional operations typically employ a combination of manual labor and scaffolding, where workers stand on movable or fixed scaffolding and climb and adjust the scaffolding height to reach the target painting position. This method is greatly affected by factors such as scaffolding stability, outdoor wind force, and limitations of high-altitude operating space. Furthermore, the paints used in exterior wall painting often contain toxic and harmful substances such as organic solvents and heavy metal ions. The painting process generates a large amount of volatile gases and pigment dust. Since exterior wall work is mostly done in open or semi-enclosed environments, long-term exposure to this environment can cause irreversible damage to workers' health. In addition, the low efficiency and inconsistent quality of manual painting are also prominent issues. On the one hand, with manual labor and scaffolding, frequent disassembly, assembly, or relocation of scaffolding is necessary to cover different painting areas, and the average daily painting area per work surface is usually insufficient. Furthermore, due to the physical exertion of workers and differences in their operational proficiency, it is difficult to maintain a uniform printing speed, which can easily lead to problems such as disordered printing rhythm in certain areas. Summary of the Invention

[0003] Therefore, it is necessary to provide a drone painting system, method, and readable storage medium to address the aforementioned technical problems.

[0004] In a first aspect, embodiments of this application provide a drone inkjet printing system, the system comprising: a perception and positioning module and an inkjet printing device installed on the drone, as well as a path planning module and a control module;

[0005] The perception and positioning module is used to acquire laser point cloud data and first image data of the area to be printed before printing, and to construct a three-dimensional model based on these data.

[0006] The path planning module is used to divide the area to be painted into grids based on the 3D model, and determine the wall feature cost, distance cost, and painting quality cost of each grid; based on the wall feature cost, distance cost, and painting quality cost of each grid, an improved method is adopted. Algorithm generates printing path;

[0007] The control module controls the flight position and attitude of the drone based on the printing path, and automatically controls the relative attitude stability between the nozzle of the printing device and the wall surface, so as to control the printing device to print evenly on the wall surface.

[0008] The sensing and positioning module is also used to acquire second image data after the first wall painting is completed;

[0009] The control module is also used to receive the second image data, determine the quality defect area based on the second image data, control the drone to fly to the quality defect area, and control the inkjet printing device to re-ink the quality defect area.

[0010] In one embodiment, the control module stores a pre-trained image recognition model, and the control module is used to input the second image data into the image recognition model and output the quality defect area;

[0011] The quality defect area includes at least unpainted blind areas, color difference areas, and uneven paint areas. The image recognition model is trained from a training image dataset, which consists of various pre-collected images of painting quality defects, including unpainted blind areas, color difference areas, and uneven paint areas. The training image dataset also includes annotation information for each quality defect, which indicates the location and type of the quality defect area in the image.

[0012] In one embodiment, the cost of the wall features is determined by at least one of the following: curvature, wall flatness, tilt angle, and obstacle distribution information.

[0013] In one embodiment, the path planning module is further configured to: establish a regional physical coordinate system based on the 3D model, identify wall geometric features through feature extraction and analysis, and perform adaptive mesh generation based on the wall geometric features; label the attributes of each mesh, and determine the wall feature cost and spraying quality cost of each mesh based on the attribute labels; calculate the distance cost between nodes in real time during path search; and adopt an improved approach based on the wall feature cost, the distance cost, and the spraying quality cost. Algorithm generates inkjet printing paths.

[0014] In one embodiment, the sensing and positioning module is further configured to: acquire real-time image data during the inkjet printing process;

[0015] The path planning module is also used to determine whether there are obstacles in the area to be painted based on the real-time image data; if so, it determines the avoidance range based on the size of the obstacle, and performs local path replanning based on the avoidance range to generate a new painting path.

[0016] The control module is also used to control the flight position and flight attitude of the UAV based on the new inkjet path.

[0017] In one embodiment, the path planning module is further configured to identify uneven areas on the wall based on the three-dimensional model, and the control module is further configured to: control the flight position of the drone based on the uneven areas on the wall, and control the length of the retractable nozzle of the inkjet printing device so that the nozzle maintains a constant distance from the wall.

[0018] In one embodiment, the system further includes: a monitoring module and an early warning module; the monitoring module is connected to the control module and is used to monitor the flight position and flight attitude of the UAV in real time.

[0019] The control module is also used to determine adjustment parameters when the flight position or flight attitude of the UAV deviates from the normal, and adjust the flight position or flight attitude of the UAV based on the adjustment parameters.

[0020] The early warning module is used to issue early warning information when the flight position or flight attitude of the UAV deviates from the normal.

[0021] In one embodiment, the sensing and positioning module is further configured to acquire third image data after the second wall painting is completed; the control module further includes an image display unit for displaying the second image data and the third image data.

[0022] Secondly, embodiments of this application also provide a method for drone inkjet printing, the method comprising:

[0023] Before printing, acquire the laser point cloud data and first image data of the area to be printed, and build a three-dimensional model based on this;

[0024] Based on the 3D model, the area to be painted is divided into meshes, and the wall feature cost, distance cost, and painting quality cost of each mesh are determined; based on the wall feature cost, distance cost, and painting quality cost of each mesh, an improved method is adopted. Algorithm generates printing path;

[0025] The flight position and attitude of the UAV are controlled based on the spraying path, and the relative attitude of the nozzle of the spraying device and the wall is stabilized so as to control the spraying device to spray the wall evenly; and the second image data after the first wall spraying is completed is acquired.

[0026] Based on the second image data, the quality defect area is determined;

[0027] Control the drone to fly to the area with the quality defect, and control the inkjet printing device to re-ink the area with the quality defect.

[0028] Thirdly, embodiments of this application also provide a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method described in the second aspect above.

[0029] The aforementioned drone-based painting system, method, and readable storage medium, through automatic control, precisely address the efficiency and full-area coverage issues of exterior wall painting: the automatic mode utilizes improved... The algorithm generates the spraying path to achieve continuous spraying, ensuring the efficiency of large-area operations and the uniformity of the coating. Combined with the posture stabilization linkage mechanism, it completes precise touch-up coating for blind spots in automatic spraying such as corners and window edges, ensuring full coverage of the exterior wall without dead angles and improving the overall quality and efficiency of rough painting operations.

[0030] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0031] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0032] Figure 1 This is a structural block diagram of a drone inkjet printing system in one embodiment;

[0033] Figure 2 This is a structural block diagram of the drone inkjet printing system in another embodiment;

[0034] Figure 3 This is a flowchart illustrating a drone painting method in one embodiment;

[0035] Figure 4 This is a flowchart illustrating a preferred embodiment of a drone printing method. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0037] Obviously, the accompanying drawings described below are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar scenarios based on these drawings without any inventive effort. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, any changes to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0038] This application provides a drone painting system that integrates an automatic control mode while retaining a manual control mode. The two modes share a common hardware foundation and collaborative control logic, enabling efficient painting operations in various scenarios. These scenarios include building exterior stone paint spraying, cleaning / firefighting, and wind turbine blade spraying and repair, providing new solutions for urban building facade painting, high-altitude glass cleaning, and wind turbine blade spraying and repair. The system relies on a closed-loop process of "environmental perception - intelligent planning - precise execution - dynamic calibration" to achieve uniform spraying across the entire exterior wall while meeting the need for precise processing of local details. Both modes share core hardware and basic modules, adapting to different scenarios only through control logic switching.

[0039] This application provides a drone-based inkjet printing system; please refer to [link / reference]. Figure 1 The diagram below illustrates the structure of the system, which includes: a perception and positioning module 10, a printing device 20, a path planning module 30, and a control module 40. The perception and positioning module 10 and the printing device 20 are mounted on the UAV. The path planning module 30 can perform its functions either through a host computer mounted on the UAV or by being located on the ground. The control module 40 is located on the ground and performs its functions.

[0040] The perception and positioning module 10 includes a lidar sensor and a vision sensor, used to acquire lidar point cloud data and first image data of the area to be painted before printing, and to construct a 3D model of the wall surface of the area to be painted based on this. The lidar point cloud data is the point cloud data collected by the lidar sensor of the perception and positioning module 10, and the image data is the image collected by the vision sensor of the perception and positioning module 10. The first image data is an image of the area to be painted collected before printing work, used for initial path planning.

[0041] The perception and positioning module 10, acting as the "perception hub," integrates multiple source sensors. Before printing, it performs a comprehensive scan of the area to be printed using a 3D laser scanner and a binocular camera to acquire laser point cloud data and first image data of the area to be printed. Based on the laser point cloud data and first image data, it constructs a 3D model of the wall surface. The 3D model of the wall surface can accurately identify the wall surface's concavity, flatness, tilt angle, and obstacle distribution, and can also identify and mark the pre-set markers for the project in advance.

[0042] The path planning module 30 is used to divide the area to be painted into grids based on the 3D model, and determine the wall feature cost, distance cost, and painting quality cost of each grid; based on the wall feature cost, distance cost, and painting quality cost of each grid, an improved method is adopted. The algorithm generates the spraying path. The path planning module 30, acting as the system's "decision-making brain," can optionally have its core functions performed by a high-performance ground control station, which relies on built-in powerful computing hardware. Alternatively, the path planning module 30 can also achieve the corresponding functions through a host computer mounted on the UAV.

[0043] The core innovation of the path planning algorithm in this application revolves around improving... Algorithm unfolding, in traditional Based on the algorithm, targeted optimizations are implemented to form a dedicated path planning logic adapted to the exterior wall painting scenario: different from traditional methods. The algorithm focuses solely on "shortest path" as its optimization objective. The path planning module 30 optimizes the path evaluation function by adding "wall feature cost" and "painting quality cost" to the traditional "distance cost." Based on the 3D wall model constructed by the perception and positioning module 10, it acquires wall flatness, obstacle distribution data, and distance data, integrating them into the evaluation system. For example, in uneven areas or around obstacles, by increasing the "feature cost weight" of local paths, the algorithm is guided to generate more precise and well-fitting detour paths that better conform to the wall's shape, avoiding the limitations of traditional methods. The algorithm is prone to problems such as "low path alignment and missed areas in spraying". Ultimately, through improvements... The path algorithm generates a serpentine spraying path from left to right and from top to bottom, with the spacing between adjacent paths matching the nozzle atomization diameter to ensure full coverage and improve work efficiency.

[0044] The control module 40 is used in automatic mode to control the flight position and attitude of the drone based on the painting path, and to stabilize the relative attitude between the nozzle of the painting device and the wall surface, so as to control the painting device to uniformly paint the wall surface. During the drone's flight, a pose adaptation algorithm is used to achieve real-time synchronization between the drone's flight trajectory and the wall curvature. This algorithm effectively counteracts the additional airflow disturbances caused by the drone approaching the wall surface, thereby maintaining the stability and uniformity of the painting process.

[0045] The sensing and positioning module 10 is also used to acquire second image data after the first wall painting is completed. A certain time interval is needed between acquiring the second image data and the first wall painting completion time; that is, the second image data needs to be acquired after the paint is completely dry to better determine whether there are quality defects such as color difference areas. The second image data is used to detect blind spots, color difference areas, and uneven paint surfaces. The sensing and positioning module 10 sends the second image data after wall painting to the control module 40. The control module receives the second image data and determines the information of the quality defect areas based on the second image data, including detailed information such as location and size. The control module controls the drone to fly to the quality defect area and controls the painting device to repaint the quality defect area. In practice, there are multiple quality defect areas. Preferably, the path planning module performs path planning based on the information of the quality defect areas to determine the optimal flight path. The control module controls the drone to fly to the quality defect area based on the optimal flight path and controls the painting device to repaint the quality defect area.

[0046] The drone painting system of this application embodiment precisely solves the problems of efficiency and full coverage in exterior wall painting through automatic control mode: the automatic mode establishes a physical coordinate system for the exterior wall and divides it into grids using grid segmentation positioning technology, and adopts an improved... The algorithm generates the spraying path to achieve continuous spraying, ensuring the efficiency of large-area operations and the uniformity of the coating. Combined with the posture stabilization linkage mechanism, it can accurately repair areas with automatic spraying quality defects such as corners and window edges, ensuring that the entire exterior wall is covered without dead angles, realizing automated spraying operations, and improving the overall quality and efficiency of spraying operations.

[0047] In one embodiment, the control module stores a pre-trained image recognition model. The control module is used to input the second image data into the image recognition model and output a quality defect region. The quality defect region includes at least unpainted blind areas, color difference areas, and uneven paint areas. The image recognition model is obtained by training a neural network on a training image dataset. The training image dataset consists of pre-collected images of various painting quality defects, including unpainted blind areas, color difference areas, and uneven paint areas. The training image dataset also includes annotation information for each quality defect, indicating the location and type of the quality defect region in the image. The training image dataset is then input into the neural network for training to obtain the trained image recognition model.

[0048] Specifically, after the first wall painting is completed, the second image data is sent to the control module. The control module then inputs the second image data into a pre-trained image recognition model, which automatically outputs areas of quality defects. This image recognition model possesses highly intelligent recognition capabilities, enabling it to quickly and accurately locate areas of quality defects upon receiving the second image data. In practical applications, once the wall painting is completed in automatic mode, the system transmits the second image data to the control module. The control module then invokes the image recognition model to quickly identify quality defects such as unpainted blind spots, color difference areas, and uneven paint areas. This process is not only efficient but also highly accurate, significantly improving the quality inspection efficiency of the painting operation and providing a reliable basis for subsequent touch-up work.

[0049] In one embodiment, the cost of the wall features is determined by at least one of the following: curvature, wall flatness, tilt angle, and obstacle distribution information.

[0050] Among them, the wall feature cost is a feature cost determined by the comprehensive multi-dimensional attributes of each grid. The multi-dimensional attributes of each grid include at least curvature, tilt angle, wall flatness, obstacle distribution information, etc.

[0051] In one embodiment, the path planning module is further configured to: establish a regional physical coordinate system based on the 3D model, identify wall geometric features through feature extraction and analysis, and perform adaptive mesh generation based on the wall geometric features; label the attributes of each mesh, and determine the wall feature cost and spraying quality cost of each mesh based on the attribute labels; calculate the distance cost between nodes in real time during path search; and adopt an improved approach based on the wall feature cost, the distance cost, and the spraying quality cost. Algorithm generates inkjet printing paths.

[0052] Specifically, after receiving the 3D wall model output by the multi-dimensional perception and positioning module 10, the path planning module 30 quickly establishes a physical coordinate system and adaptively divides the area to be painted based on the 3D model, forming a non-uniform grid distribution. The grid is adaptively divided according to the wall's geometric features, resulting in a non-uniform grid. Larger grids are used in flat areas, while smaller grids are used in uneven areas, edge corners, and around obstacles, thus creating a non-uniform grid. Then, based on the divided grid, the wall feature cost and the painting quality cost based on the painting difficulty and coverage uniformity requirements are determined for each grid. During the path search process, search nodes are created for the grids to be expanded, and the distance cost between nodes is calculated in real time, thereby determining the three costs of the grid: wall feature cost, painting quality cost, and distance cost. The weights of the three costs are dynamically allocated according to whether the grid where the node is located is a flat area or a complex uneven area, thus achieving adaptive path planning that prioritizes efficiency in flat areas and quality in complex areas.

[0053] The path planning module in this application adopts an improved method. The algorithm generates the optimal printing path, an improvement. The improvement of the algorithm compared to the traditional algorithm lies in the introduction of a multi-objective cost function. The multi-objective cost function is constructed based on the wall feature cost, spraying quality cost and distance cost. This function comprehensively considers the geometric distance of the path, the influence of the physical features of the wall on the motion, and the quality requirements of the spraying process.

[0054] The wall feature cost is a comprehensive feature cost determined based on the multi-dimensional attributes of each grid. These attributes include at least curvature, tilt angle, flatness, and obstacle distribution information. For example, the greater the curvature of the grid's inner surface, the higher the feature cost; the greater the tilt angle of the grid surface, the higher the feature cost; the worse the flatness (unevenness), the higher the feature cost; and the closer to an obstacle, the higher the feature cost. The distance cost is a feature cost determined based on the geometric distance of the actual path from the starting node to the current node. The spraying quality cost is a feature cost determined based on the spraying coverage difficulty and quality requirements of the grid containing the current node.

[0055] In one embodiment, the perception and positioning module 10 is further configured to acquire real-time image data during the printing process; the real-time image data is used for dynamic planning in the UAV printing operation to cope with dynamic obstacles and environmental changes; the path planning module 30 is further configured to determine whether there are obstacles in the area to be printed based on the real-time image data; if so, the avoidance range is determined based on the size of the obstacle, and local path replanning is performed based on the avoidance range to generate a new printing path; the control module 40 is further configured to control the flight position and flight attitude of the UAV based on the new printing path.

[0056] The path planning module of this application, when detecting new obstacles in real time during path execution, improves upon the following: The algorithm can trigger local replanning from the current position, ensuring the continuity and integrity of the operation.

[0057] The path planning module 30 in this embodiment can perform dynamic replanning response, and perform local path replanning for temporary obstacles encountered during the exterior wall painting process (such as sudden pipelines or wall protrusions). Improved The algorithm abandons the inefficient traditional "full path recalculation" mode. By pre-setting a "local path buffer," it quickly recalculates only the path segments within the obstacle's influence range. Simultaneously, it links with the attitude stabilization system data to correct path nodes, ensuring that the replanning process does not interrupt the spraying operation. This perfectly adapts to the collaborative needs of "continuous and efficient spraying" and "precise recoating of quality defect areas (blind spots, color difference areas, and uneven paint areas)" in the automatic mode of the device. This improvement allows path planning to retain the traditional... The algorithm's path optimization efficiency also solves its shortcomings of "weak adaptability and slow response" in dynamic and complex exterior wall scenarios, achieving "full coverage without dead angles + efficient spraying".

[0058] In one embodiment, the path planning module 30 is further configured to identify uneven areas on the wall based on the three-dimensional model, and the control module is further configured to: control the flight position of the drone based on the uneven areas on the wall, and control the length of the retractable nozzle of the inkjet printing device so that the nozzle maintains a constant distance from the wall.

[0059] Specifically, the spraying device 20 includes a paint supply device, an air pressure device, and a nozzle length adjustment device. To accommodate spraying of irregularly shaped and ultra-high structures, an onboard paint tank is selected as the paint supply storage method. The onboard paint tank and the UAV frame are connected by a quick-release structure, ensuring connection stability while providing the possibility of rapid tank replacement during actual construction, thus ensuring construction continuity and improving efficiency. Considering the UAV's payload, the air pressure device preferentially uses a separate air pump to drive the spraying system. When the UAV's load has redundancy, a pressure tank or similar device can be used to provide higher pressure. To accommodate the spraying needs of various paints, the maximum air pressure output of the air pump is not less than 0.4 MPa.

[0060] In automatic mode, the inkjet printing device 20 regulates the paint output through the intelligent spray gun according to control commands. When encountering uneven areas on the wall during the inkjet printing process, the distance data between the nozzle and the wall is measured by a laser sensor and sent to the control module 40. The control module 40 receives the distance data and controls the movement of the telescopic nozzle based on the distance data to adjust the length of the telescopic nozzle so that the nozzle and the wall maintain a constant distance, automatically compensating for unevenness on the wall. At the same time, the attitude stabilization system is used to calibrate the attitude of the drone in real time to ensure that the relative attitude between the nozzle and the wall is stable during the inkjet printing process, thereby achieving uniform coating.

[0061] In one embodiment, such as Figure 2 As shown, the system also includes a monitoring module 50, which is connected to the control module and is used to monitor the flight position and flight attitude of the UAV in real time. The control module is also used to determine adjustment parameters when the flight position or flight attitude of the UAV deviates from the normal, and to adjust the flight position or flight attitude of the UAV based on the adjustment parameters.

[0062] The monitoring module 50 is used to synchronize key data such as spraying progress, attitude parameters, and material balance in real time. Its core function is dynamic monitoring and closed-loop control: the attitude sensing module continuously captures the UAV's tilt angle, sway amplitude, and other states. When attitude fluctuations exceed the threshold or distance fluctuations are abnormal, a dual mechanism is immediately triggered: the spraying function is automatically locked to avoid paint waste or defects, while abnormal flight data is returned to the ground in real time, and the attitude stabilization system is linked to quickly calibrate the flight attitude. The monitoring module 50 is also used to monitor the remaining paint in the paint bucket.

[0063] In one embodiment, such as Figure 2 As shown, the system also includes an early warning module 60, which is used to issue early warning information when the flight position or flight attitude of the UAV deviates from the normal.

[0064] The early warning module 60 is used to issue early warning information when the UAV's flight position or attitude deviates from the normal range. Simultaneously, when insufficient material is detected, it automatically issues an early warning and triggers stop-spray and hover commands to ensure operational safety. All abnormal data and calibration process data are synchronously uploaded to the cloud, supporting subsequent quality backtracking analysis and providing a basis for attitude stabilization algorithm optimization. This forms a complete "monitoring-early warning-calibration-recording" process, ensuring the stability and accuracy of automated spraying operations.

[0065] In one embodiment, the sensing and positioning module is further configured to acquire third image data after the second wall painting is completed; the control module further includes an image display unit for displaying the second image data and the third image data, allowing the user to perform painting quality detection and evaluation based on the third image data.

[0066] After determining the quality defect area based on the second image data, the control module 40 controls the drone to fly to the quality defect area. At the same time, the image display unit is also used to display the collected real-time image data. While the control module 40 automatically controls the drone and the inkjet printing device, the user can view the real-time image data and make corresponding decisions through the real-time image data. The user can also manually execute corresponding control commands to control the drone and the inkjet printing device to re-ink the quality defect area to make up for minor defects in the automatic control mode.

[0067] In this embodiment, when repainting areas with quality defects, a manual control mode can be used. In manual control mode, the focus is mainly on complex area operations. Through the "human-machine collaboration + attitude stabilization linkage" mechanism, the drone operator directly controls the drone's position and painting commands using the control input module (Hall effect universal joint, painting trigger button, high-definition image transmission system) on the remote controller. The system calls the attitude perception and stabilization module to compensate for attitude deviations in real time (ensuring that the nozzle is perpendicular to the wall). The laser distance sensor feeds the distance data back to the display screen to help maintain a stable spray distance. The painting execution module adjusts the nozzle parameters according to the manual commands to complete manual wall painting and repainting of areas with quality defects in automatic mode.

[0068] This application also provides a method for drone inkjet printing, such as... Figure 3 As shown, the method includes the following steps:

[0069] Step S201: Before printing, acquire the laser point cloud data and first image data of the area to be printed, and construct a three-dimensional model based on them;

[0070] Step S202: Based on the 3D model, the area to be painted is meshed, and the wall feature cost, distance cost, and painting quality cost are determined; based on the wall feature cost, the distance cost, the painting quality cost, and the meshed area to be painted, an improved method is adopted. Algorithm generates printing path;

[0071] Step S203: Control the flight position and attitude of the drone based on the printing path, and control the relative attitude of the nozzle of the printing device and the wall to stabilize, so as to control the printing device to print evenly on the wall; and acquire the second image data after the first wall printing is completed.

[0072] Step S204: Based on the second image data, determine the quality defect area;

[0073] Step S205: Control the drone to fly to the quality defect area, and control the inkjet printing device to re-ink the quality defect area.

[0074] Figure 4 This is a preferred flowchart of the drone inkjet printing method in this embodiment, as follows: Figure 4 As shown, this drone painting method combines automatic and manual modes, and specifically includes the following steps:

[0075] The intelligent inkjet printing drone is set to automatic control mode. A designated painting area is set, and the sensing and positioning module on the drone provides laser point cloud data and image data of the painting area, which is sent to the path planning module 30 on the ground. The path planning module 30 generates the painting path and corresponding path instructions. The control module 40 receives the path instructions and controls the drone to fly to the designated position. Before starting the inkjet printing device, it checks whether the drone's flight attitude is stable. If unstable, it returns abnormal data to the ground station, which provides calibration data to the control module 40 for calibration. When the drone's flight attitude is found to be stable, the control module starts the inkjet printing device to begin printing. During printing, the high-definition camera of the sensing and positioning module 10 acquires the printed image in real time and sends it to the display unit on the control module 40 via an image transmitter.

[0076] The intelligent inkjet printing drone is set to manual control mode. Based on the printed image displayed on the display unit, the drone is controlled to fly to the positioning position by manually operating the control module 40. Before starting the inkjet printing device, the drone's flight attitude is checked to see if it is stable. When the drone's flight attitude is found to be stable, the inkjet printing device is controlled to start printing.

[0077] This application also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the UAV painting method as described in the above embodiments.

[0078] It should be noted that the steps shown in the above process or in the flowchart of the accompanying figures can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0079] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in any of the above embodiments of the UAV inkjet printing system.

[0080] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0081] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] The embodiments described above are merely illustrative of several implementation methods of this application and should not be construed as limiting the scope of the invention patent. For those skilled in the art, various modifications and improvements can be made without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A drone inkjet printing system, characterized in that, The system includes: a perception and positioning module and a printing device installed on the UAV, as well as a path planning module and a control module; The perception and positioning module is used to acquire laser point cloud data and first image data of the area to be printed before printing, and to construct a three-dimensional model based on these data. The path planning module is used to divide the area to be painted into grids based on the 3D model, and determine the wall feature cost, distance cost, and painting quality cost of each grid; based on the wall feature cost, distance cost, and painting quality cost of each grid, an improved method is adopted. Algorithm generates printing path; The control module is used to control the flight position and attitude of the drone based on the printing path, and to control the relative attitude stability between the nozzle of the printing device and the wall surface, so as to control the printing device to print evenly on the wall surface. The sensing and positioning module is also used to acquire second image data after the first wall painting is completed; The control module is also used to receive the second image data, determine the quality defect area based on the second image data, control the drone to fly to the quality defect area, and control the inkjet printing device to re-ink the quality defect area.

2. The system according to claim 1, characterized in that, The control module stores a pre-trained image recognition model. The control module is used to input the second image data into the image recognition model and output the quality defect area. The quality defect area includes at least unpainted blind areas, color difference areas, and uneven paint areas. The image recognition model is trained from a training image dataset, which consists of various pre-collected images of painting quality defects, including unpainted blind areas, color difference areas, and uneven paint areas. The training image dataset also includes annotation information for each quality defect, which indicates the location and type of the quality defect area in the image.

3. The system according to claim 1, characterized in that, The cost of the wall features is determined by at least one of the following: curvature, wall flatness, tilt angle, and obstacle distribution information.

4. The system according to claim 1, characterized in that, The path planning module is further configured to: establish a regional physical coordinate system based on the 3D model, identify wall geometric features through feature extraction and analysis, and perform adaptive mesh generation based on the wall geometric features; label the attributes of each mesh, and determine the wall feature cost and spraying quality cost of each mesh based on the attribute labels; calculate the distance cost between nodes in real time during path search; and adopt an improved approach based on the wall feature cost, the distance cost, and the spraying quality cost. Algorithm generates inkjet printing paths.

5. The system according to claim 1, characterized in that, The sensing and positioning module is also used to: acquire real-time image data during the inkjet printing process; The path planning module is also used to determine whether there are obstacles in the area to be painted based on the real-time image data; if so, it determines the avoidance range based on the size of the obstacle, and performs local path replanning based on the avoidance range to generate a new painting path. The control module is also used to control the flight position and flight attitude of the UAV based on the new inkjet path.

6. The system according to claim 1, characterized in that, The path planning module is also used to identify uneven areas on the wall based on the three-dimensional model. The control module is also used to control the flight position of the drone based on the uneven areas on the wall, and to control the length of the retractable nozzle of the inkjet printing device so that the nozzle maintains a constant distance from the wall.

7. The system according to claim 1, characterized in that, The system also includes a monitoring module and an early warning module; the monitoring module is connected to the control module and is used to monitor the flight position and flight attitude of the UAV in real time. The control module is also used to determine adjustment parameters when the flight position or flight attitude of the UAV deviates from the normal, and adjust the flight position or flight attitude of the UAV based on the adjustment parameters. The early warning module is used to issue early warning information when the flight position or flight attitude of the UAV deviates from the normal.

8. The system according to claim 1, characterized in that, The sensing and positioning module is also used to acquire third image data after the second wall painting is completed; the control module further includes an image display unit for displaying the second image data and the third image data.

9. A method for printing inkjet prints using a drone, characterized in that, The method includes: Before printing, acquire the laser point cloud data and first image data of the area to be printed, and build a three-dimensional model based on this; Based on the 3D model, the area to be painted is divided into meshes, and the wall feature cost, distance cost, and painting quality cost of each mesh are determined; based on the wall feature cost, distance cost, and painting quality cost of each mesh, an improved method is adopted. Algorithm generates printing path; The flight position and attitude of the UAV are controlled based on the spraying path, and the relative attitude of the nozzle of the spraying device and the wall is stabilized so as to control the spraying device to spray the wall evenly; and the second image data after the first wall spraying is completed is acquired. Based on the second image data, the quality defect area is determined; Control the drone to fly to the area with the quality defect, and control the inkjet printing device to re-ink the area with the quality defect.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of claim 9.

Images