A cleaning application and data processing system suitable for drones

Abstract

The present invention relates to the field of drone cleaning technology and discloses a cleaning application and data processing system suitable for drones. The system includes a stain distribution modeling module, a sensor configuration module, a task definition module, an execution control module, and an analysis and feedback module. The system constructs a stain distribution model through multispectral imaging, arranges detection points to calculate the stain density gradient, defines cleaning indicators based on physical parameters, performs steady-state, dynamic response, and composite cleaning, and performs abnormality tracing and strategy updates for areas that fail cleaning. In addition, the system also includes data aggregation and communication relay modules. The system can accurately model and scientifically control cleaning parameters, adapt to complex working conditions, improve cleaning efficiency and quality, and is suitable for the field of drone cleaning.

Description

Technical Field

[0001] The present invention relates to the technical field of drone cleaning, and in particular to a cleaning application and data processing system suitable for drones. Background Art

[0002] With the widespread use of drones in various fields, their surfaces are easily stained during operation, affecting the performance and service life of the drone. Currently, there are many problems with drone cleaning.

[0003] When it comes to cleaning planning, traditional cleaning methods lack the ability to accurately model the distribution of stains on drone surfaces. Due to the complex structure of drone surfaces, the types and levels of stains found in different areas vary significantly. Failure to tailor cleaning strategies based on stain distribution can lead to inefficient cleaning, either over-cleaning certain areas and wasting resources, or under-cleaning others, impacting cleaning effectiveness.

[0004] From a sensor configuration perspective, existing cleaning systems lack a scientific basis for the placement of detection points on drone surfaces. This inability to accurately capture information such as stain density across the drone's surface makes it difficult to adjust cleaning parameters, such as water pressure and detergent ratio, based on actual conditions, leading to inconsistent cleaning results.

[0005] When it comes to cleaning, traditional systems often use a single cleaning mode, such as steady-state cleaning with a fixed water pressure and detergent ratio. This mode cannot adapt to the cleaning needs of different areas on the drone's surface, especially areas with high levels of stains, where a single cleaning mode often fails to effectively remove them. Furthermore, when faced with complex working conditions, traditional systems lack dynamic response capabilities and cannot adjust cleaning parameters in a timely manner, affecting cleaning quality.

[0006] Data processing and feedback mechanisms are also inadequate. Existing systems don't fully collect and analyze data during the cleaning process, making it impossible to accurately trace abnormalities to uncleaned areas. This makes it difficult to generate effective diagnostic logs and update cleaning strategies, hindering the improvement and optimization of cleaning technology.

[0007] Furthermore, drones operate in complex and variable environments during flight, such as varying weather conditions, flight speeds, and altitudes, all of which affect the adhesion and removal of stains. Traditional cleaning systems lack the ability to simulate and process cleaning responses under these varying operating conditions, resulting in poor adaptability. Summary of the Invention

[0008] The purpose of the present invention is to provide a cleaning application and data processing system suitable for drones to solve the problems raised in the above background technology.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a cleaning application and data processing system suitable for drones, the system comprising:

[0010] Stain distribution modeling module: Scans the target cleaning surface structure to generate a stain distribution model, constructs a spatial coordinate system with the geometric center of the stain distribution model, uses the stain distribution model as a reference model, and uses the spatial coordinate system as a path planning framework to create reference points of the reference model at preset intervals. The reference points include basic stain points, key stain points, and matching stain points.

[0011] Sensor configuration module: Defines the surface to be cleaned as a work target, places the work target in the path planning framework, arranges detection points of the work target at preset intervals, including basic detection points, key detection points, and matching detection points, aligns the benchmark model with the work target, and defines the difference in stain density between the detection point and the benchmark point as the stain density gradient;

[0012] Task definition module: obtains the physical parameters of the target surface and defines preset cleaning indicators based on the physical parameters;

[0013] Execution control module: performs cleaning operations on the job target, including steady-state cleaning, dynamic response cleaning, and composite cleaning. Dynamic response cleaning includes water pressure regulation, detergent ratio adjustment, and spray angle adjustment.

[0014] Analysis and feedback module: Trace the anomalies of job targets that fail dynamic response cleaning, generate diagnostic logs and update cleaning strategies.

[0015] Preferably, the stain distribution modeling module includes:

[0016] Scan the target cleaning surface through multispectral imaging to build a stain distribution model;

[0017] A spatial coordinate system is established based on the geometric center of the stain distribution model, and reference points are generated, including basic stain points, key stain points, and matching stain points. The reference points evenly cover the stain distribution model and the distance between adjacent reference points is equal.

[0018] The reference points in the high-adhesion stain area are used as key stain points, and the reference points in the axial three-divided sections of the work target are used as matching stain points. The coordinate set of the reference points is recorded. , is the number of benchmark points;

[0019] The stain distribution model is used as the benchmark model and the spatial coordinate system is used as the path planning framework.

[0020] Preferably, the sensor configuration module includes:

[0021] Define the surface to be cleaned as the work target, locate the work target in the path planning framework, and arrange inspection points including basic inspection points, key inspection points and matching inspection points. The inspection points evenly cover the work target and the distance between adjacent inspection points is equal.

[0022] The detection points in the high-adhesion stain area of ​​the work target are used as key detection points, and the detection points in the axial three-divided sections of the work target are used as matching detection points. The coordinate set of the detection points is recorded. , is the number of detection points;

[0023] Overlapping the matching benchmark points with the matching detection points to achieve alignment between the work target and the benchmark model;

[0024] The difference in stain density between the detection point and the reference point is defined as the stain density gradient , including the base point gradient , key point gradient , matching point gradient ;

[0025] Get the number of benchmark points and the number of testing points ,definition and The ratio is the stain point missing rate .

[0026] Preferably, the task definition module includes:

[0027] Obtaining physical parameters of the target surface including surface roughness, stain adhesion, maximum water pressure limit, and cleaning agent chemical activity threshold;

[0028] The preset cleaning indicators are defined according to the physical parameters, including the preset stain density gradient threshold, the preset stain point missing rate threshold, and the preset water pressure adjustment threshold.

[0029] Preferably, the execution control module includes:

[0030] Perform steady-state cleaning, dynamic response cleaning, and composite cleaning on the work targets located in the path planning framework;

[0031] The steps of steady-state cleaning are: obtaining the coordinates and number of detection points and reference points, calculating the stain density gradient and stain point missing rate ,like and If the preset dirt density gradient threshold and the preset dirt point missing rate threshold are met at the same time, the operation target is determined to have passed the steady-state cleaning;

[0032] A dynamic response cleaning environment is constructed for the steady-state cleaning operation target, and water pressure, detergent ratio and spray uniformity are adjusted. The adjustment results are output in a quantitative matrix.

[0033] Composite cleaning is performed on the job targets through dynamic response cleaning to simulate the cleaning response behavior under variable working conditions.

[0034] Preferably, the steps of constructing a dynamic response cleaning environment and outputting a quantization matrix are:

[0035] Define dynamic response factors for job objectives Including water pressure response factor , detergent response factor , spray uniformity factor ;

[0036] The water pressure response factor Including high adhesion area water pressure factor and low adhesion area water pressure factor, spray uniformity factor Including axial uniformity factor and radial uniformity factor;

[0037] Build a dynamic response cleaning environment, including step-by-step water pressure instructions and step-by-step detergent concentration instructions applied to the work target;

[0038] The water pressure response test result is defined as the water pressure rate in the high adhesion area and water pressure rate in the low adhesion area , the spray uniformity test result is axial deviation and radial deviation ;

[0039] like 、 、 、 If the corresponding preset water pressure regulation threshold and the preset spray uniformity threshold are met, it is determined that the operation target has passed the dynamic response cleaning.

[0040] Preferably, the steps of performing composite cleaning are:

[0041] The comprehensive cleaning stability coefficient of the calculated operation target is:

[0042]

[0043] in, is the comprehensive cleaning stability coefficient, is the number of variable working condition scenarios, and is the weight coefficient, Indicates the The dynamic response factor under a variable working condition scenario, Indicates the The spray uniformity factor under different working conditions.

[0044] Preferably, the analysis and feedback module includes:

[0045] Obtain cleaning data for targets that fail dynamic response cleaning and calculate the direction vector of the stain density gradient ,in is the coordinate vector of the reference point, is the detection point coordinate vector;

[0046] based on Analyze the conduction path of residual stains and locate cleaning blind spots or high residue areas;

[0047] The work target is divided into a finite number of equal-thickness sections along the axial direction, a preset detection point density threshold is defined, the number of detection points in each section is counted, the section detection point density is calculated, and the surface roughness verification is performed on sections that exceed the preset threshold.

[0048] Preferably, the system further comprises:

[0049] The data aggregation module is connected with the stain distribution modeling module and the sensor configuration module to periodically obtain the reference point coordinate set and the detection point coordinate set , construct the stain gradient distribution matrix, store the stain gradient distribution matrix in the local database, where the data aggregation cycle is synchronized with the execution cycle of dynamic response cleaning, the dimension of the stain gradient distribution matrix and the number of reference points consistent.

[0050] Preferably, the system further comprises:

[0051] The communication relay module is connected to the external flight control system, encapsulates the adjustment results of the execution control module and the diagnostic log of the analysis and feedback module into a flight control protocol data packet, and transmits it to the flight control system through a wireless network.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The system utilizes a stain distribution modeling module, using multispectral imaging to scan the target cleaning surface and construct a stain distribution model. This model then establishes a spatial coordinate system based on the geometric center, generating a reference point containing basic, critical, and matching stain points. This uniformly covers the stain distribution model, laying the foundation for precise path planning and cleaning strategy development. This precise modeling approach allows the system to clearly understand the distribution of stains on the drone surface, including key areas with high stain adhesion, allowing for targeted cleaning, improving cleaning efficiency, and avoiding resource waste and insufficient cleaning.

[0054] The sensor configuration module defines the surface to be cleaned as the target and locates it within the path planning framework. It then arranges detection points corresponding to the reference points. By calculating the stain density gradient and stain point loss rate, the target is aligned with the reference model, providing accurate data support for subsequent adjustments to cleaning parameters. This scientific detection point layout and data calculation enable the system to accurately and in real time acquire stain information from all areas of the drone's surface, providing a reliable basis for dynamically adjusting cleaning parameters and ensuring a more precise and effective cleaning process.

[0055] The task definition module captures the physical parameters of the target surface, such as surface roughness and stain adhesion, and defines preset cleaning indicators accordingly. This gives the cleaning task clear standards and objectives, ensuring that the cleaning results meet the requirements. Clear cleaning indicators further standardize the cleaning process and ensure the stability and consistency of cleaning quality.

[0056] The execution control module can perform steady-state cleaning, dynamic response cleaning, and composite cleaning. Steady-state cleaning determines pass / fail by judging the stain density gradient and the stain point loss rate, ensuring basic cleaning effectiveness. Dynamic response cleaning allows for adjustments to water pressure, detergent ratio, and spray angle, allowing for flexible adjustments to different stain conditions and improved cleaning effectiveness. Composite cleaning simulates cleaning response behavior under variable operating conditions, enhancing the system's adaptability in complex working conditions. The combination of multiple cleaning modes enables the system to cope with a variety of complex stain conditions on drone surfaces and diverse working environments, greatly improving the flexibility and adaptability of the cleaning system and ensuring excellent cleaning results in all situations.

[0057] The analysis and feedback module traces the source of abnormalities for targets that fail dynamic response cleaning, locates blind spots or high-residue areas, verifies surface roughness, generates diagnostic logs, and updates cleaning strategies, all of which contribute to system optimization and improvement. Timely abnormality tracing and strategy updates enable the system to continuously learn and improve, enhancing cleaning efficiency and quality and ensuring the long-term stability of drone operations.

[0058] The data aggregation module periodically acquires coordinates of reference points and inspection points, constructs and stores a soil gradient distribution matrix, and provides data support for subsequent analysis and optimization. This helps identify patterns and issues in the cleaning process and further improves system performance. This rich data accumulation and analysis provides strong support for the continuous improvement and upgrade of the system, enabling it to continuously adapt to new needs and challenges.

[0059] The communication relay module connects to the external flight control system, transmitting adjustment results and diagnostic logs. This enables the system to work collaboratively with the drone's flight control, improving the overall efficiency and reliability of drone operations. This collaborative operation enables more coordinated drone flight and cleaning operations, improving the efficiency and reliability of drones and providing better support for their widespread application. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a working principle diagram of the cleaning application and data processing system for drones according to the present invention;

[0061] Figure 2 The working principle diagram of the stain distribution modeling module;

[0062] Figure 3 Module workflow diagram for sensor configuration;

[0063] Figure 4 Build and adjust flow charts for dynamically responsive cleaning environments. DETAILED DESCRIPTION

[0064] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0065] See also Figures 1-4 The present invention relates to a cleaning application and data processing system suitable for drones. The system includes: a stain distribution modeling module, a sensor configuration module, a task definition module, an execution control module, and an analysis and feedback module. The specific implementation steps are as follows:

[0066] Stain distribution modeling module: Scans the target cleaning surface structure to generate a stain distribution model, constructs a spatial coordinate system with the geometric center of the stain distribution model, uses the stain distribution model as a reference model, and uses the spatial coordinate system as a path planning framework to create reference points of the reference model at preset intervals. The reference points include basic stain points, key stain points, and matching stain points.

[0067] Sensor configuration module: Defines the surface to be cleaned as a work target, places the work target in the path planning framework, arranges detection points of the work target at preset intervals, including basic detection points, key detection points, and matching detection points, aligns the benchmark model with the work target, and defines the difference in stain density between the detection point and the benchmark point as the stain density gradient;

[0068] Task definition module: obtains the physical parameters of the target surface and defines preset cleaning indicators based on the physical parameters;

[0069] Execution control module: performs cleaning operations on the job target, including steady-state cleaning, dynamic response cleaning, and composite cleaning. Dynamic response cleaning includes water pressure regulation, detergent ratio adjustment, and spray angle adjustment.

[0070] Analysis and feedback module: Trace the anomalies of job targets that fail dynamic response cleaning, generate diagnostic logs and update cleaning strategies.

[0071] The present invention will be further described below in conjunction with Examples 1 to 5:

[0072] Example 1:

[0073] The implementation method of the stain distribution modeling module in this embodiment is as follows: The core function of this module is to scan the target cleaning surface using multispectral imaging technology to construct a stain distribution model, and based on this, establish a path planning framework and generate reference points. The specific working process is to first use a multispectral imaging device to perform a scanning operation on the target cleaning surface. The multispectral imaging device can emit electromagnetic waves of different wavelengths, receive signals reflected or radiated by the target surface, and then obtain image data containing rich spectral information. This image data can accurately reflect the characteristics of stains in different areas of the target surface, such as stain type, color, texture, and distribution status. For example, different stains such as oil, dust, and mud will show unique spectral characteristics in the multispectral image. Based on this, the stains can be identified and distinguished, laying the foundation for building an accurate stain distribution model.

[0074] After acquiring multispectral image data, the data is processed and analyzed to construct a stain distribution model. This model digitally describes the stain distribution on the target surface, integrating information such as the spatial location, range, and density of the stains into a visual model, allowing operators to intuitively understand the stain condition of the target surface.

[0075] Next, based on the constructed stain distribution model, its geometric center is determined. This can be determined by calculating the average coordinates of all stain points in the model, or by using other appropriate geometric calculation methods to ensure that the determined geometric center accurately represents the spatial center of the stain distribution model. A spatial coordinate system is established with this geometric center as the origin. This spatial coordinate system comprises x-, y-, and z-axes, corresponding to the length, width, and height of the target surface, respectively. This provides a unified spatial reference framework for subsequent path planning and cleaning operations.

[0076] After establishing the spatial coordinate system and the reference model, create reference points on the reference model at a preset spacing. The preset spacing should be determined based on factors such as the target surface size, the complexity of the stain distribution, and the required cleaning precision. For example, for target surfaces with relatively uniform stain distribution and low cleaning precision requirements, the preset spacing can be increased appropriately. For target surfaces with complex stain distribution and high cleaning precision requirements, the preset spacing should be decreased to ensure that the reference points fully cover the stain distribution model and capture sufficient stain information.

[0077] The reference points include basic stain points, key stain points, and matching stain points. Basic stain points evenly cover the entire stain distribution model, and the spacing between adjacent basic stain points is equal to achieve comprehensive sampling of the stain distribution. Key stain points are selected as reference points in high-adhesion stain areas. High-adhesion stain areas generally refer to areas where stains are tightly attached to the target surface and difficult to clean, such as areas where oil stains have been attached for a long time and stains on uneven surfaces. Cleaning stains in these areas is more difficult and requires special attention in subsequent cleaning operations. Therefore, their corresponding reference points are set as key stain points to allow for special treatment during path planning and cleaning parameter adjustment.

[0078] The matching stain point serves as the reference point for the axial trisection of the target. Divide the target axially (i.e., along a specific axis of the spatial coordinate system, such as the z-axis) into three equal sections, resulting in two equal sections. The reference points selected on these two sections are the matching stain points. The purpose of setting the matching stain point is to subsequently align the reference model with the target in the sensor configuration module. By aligning the matching reference point with the matching detection point on the target, the spatial consistency of the two is ensured, providing an accurate spatial reference for subsequent stain density gradient calculations and cleaning operations.

[0079] In the process of generating reference points, the coordinates of each reference point need to be accurately recorded. The reference point coordinate set is expressed as ,in The number of reference points. The coordinates must be recorded based on the established spatial coordinate system to ensure that the position of each reference point is unique and accurate in space.

[0080] Finally, the constructed stain distribution model serves as a baseline model, and the established spatial coordinate system serves as a path planning framework. The baseline model provides a reference standard for stain distribution for subsequent cleaning operations, while the path planning framework provides the spatial basis for planning the drone's flight path and cleaning operations. Based on the baseline model and path planning framework, the drone can develop a reasonable cleaning path and strategy, ensuring efficient and accurate cleaning operations.

[0081] Example 2:

[0082] This embodiment describes in detail the implementation method of the sensor configuration module, which is as follows: The core function of this module is to define the surface to be cleaned as the work target and locate it in the path planning framework, and provide data support for subsequent cleaning operations by laying out detection points, aligning the reference model with the work target, and calculating the stain density gradient and stain point missing rate. The specific working process is to first clarify the scope and boundaries of the surface to be cleaned and define it as the work target. The surface to be cleaned can be different parts of the drone, such as the wings, fuselage, propellers, etc. There are differences in the shapes and structures of different parts. Therefore, when defining the work target, it is necessary to accurately define it according to the characteristics of the actual parts to be cleaned to ensure the accuracy of the work target.

[0083] Positioning technology is used to precisely locate the work target within the path planning framework established by the stain distribution modeling module. During positioning, the spatial coordinate system of the path planning framework is referenced to determine the specific position and posture of the work target within that coordinate system. For example, by obtaining the coordinates of several feature points on the work target and matching them with the coordinates in the path planning framework, the work target is precisely located, ensuring that the work target's position in the path planning framework is consistent with its actual position, providing an accurate spatial reference for subsequent inspection point placement and cleaning operations.

[0084] After the target is located, inspection points are placed on the target at preset spacings. This spacing corresponds to the preset spacing of the reference points in the soil distribution modeling module to ensure consistency in the spatial sampling density between the inspection points and the reference points, facilitating subsequent data comparison and analysis. Inspection point placement must adhere to the principle of uniform coverage, meaning that inspection points are evenly distributed across the target surface, with equal spacing between adjacent inspection points. This ensures that every area of ​​the target surface can be effectively inspected and avoids blind spots.

[0085] Inspection points include basic inspection points, key inspection points, and matching inspection points. Basic inspection points evenly cover the surface of the work target and are used to conduct a comprehensive, general inspection of the stain condition on the work target surface. Key inspection points are selected from inspection points in areas of high stain adhesion on the work target. High stain adhesion areas are determined based on the identification results of high stain adhesion areas in the stain distribution modeling module. These areas are difficult to clean and require focused monitoring. Therefore, the corresponding inspection points are designated as key inspection points, allowing for more detailed attention and treatment of these areas during subsequent cleaning processes.

[0086] Matching checkpoints are points that divide the target's axial section into three equal parts. Corresponding to the matching reference points in the benchmark model, the target's axial direction (e.g., along the z-axis of the path planning framework) is divided into three equal parts, resulting in two cross-sections. Checkpoints are placed on these two cross-sections, and these checkpoints are known as matching checkpoints. The purpose of setting matching checkpoints is to achieve precise alignment between the benchmark model and the target. By aligning matching reference points with matching checkpoints, the spatial consistency of the two is ensured, laying the foundation for subsequent calculation of the stain density gradient.

[0087] In the process of arranging detection points, it is necessary to accurately record the coordinates of each detection point. The detection point coordinate set is expressed as ,in is the number of detection points. The coordinates are recorded based on the spatial coordinate system of the path planning framework to ensure the uniqueness and accuracy of the position of each detection point in space, which facilitates subsequent comparison and calculation with the coordinates of the reference points.

[0088] After completing the layout and coordinate recording of the checkpoints, the reference model is aligned with the work target. This alignment process involves aligning the matching reference points with the matching checkpoints. By adjusting the position and posture of the work target within the path planning framework, the coordinates of the two matching points in the spatial coordinate system are completely consistent. This alignment process requires precise positioning and adjustment techniques to ensure accurate spatial alignment between the reference model and the work target, thus preventing deviations in the subsequent calculation of the stain density gradient due to alignment errors.

[0089] After alignment is completed, the difference in stain density between the detection point and the reference point is defined as the stain density gradient. The stain density gradient includes the base point gradient , key point gradient , matching point gradient . Basic point gradient The difference in stain density between the basic detection point and the corresponding basic reference point is used to reflect the change in stain density in the general area of ​​the target surface; the key point gradient The difference in stain density between the key detection point and the corresponding key reference point is used to reflect the change in stain density in the high-adhesion stain area; the matching point gradient It is the difference in stain density between the matching detection point and the corresponding matching reference point, which is used to verify the consistency of stain density after the reference model is aligned with the work target.

[0090] When calculating the stain density gradient, it is necessary to first obtain the stain density value of each detection point and the corresponding reference point. The stain density value can be obtained through sensor detection technology, such as using image recognition technology to analyze the image of the area where the detection point is located and calculate the stain density of the area. The stain density gradient is obtained by subtracting the stain density value of the detection point from the stain density value of the reference point. ,The gradient value can intuitively reflect the difference in stain distribution between the target surface and the ,reference model, and provide a basis for the subsequent cleaning strategy ,formulation.

[0091] In addition, get the number of benchmark points and the number of testing points ,calculate and The ratio of the stain point missing rate is defined as . Stain point missing rate It is used to measure the difference between the number of detection points on the target surface and the number of reference points on the benchmark model. In actual applications, due to the shape, size or detection conditions of the target, the number of detection points may be inconsistent with the number of reference points. By calculating the missing rate of stain points , the integrity and effectiveness of the detection point can be evaluated. If it is too high, it may mean that the layout of the detection points is not comprehensive enough, and the layout plan of the detection points needs to be readjusted to ensure more accurate and comprehensive stain detection on the target surface.

[0092] The implementation process of the entire sensor configuration module, through a series of operations such as work target definition and positioning, detection point layout, alignment of the benchmark model with the work target, and calculation of the stain density gradient and stain point missing rate, achieves comprehensive detection and data collection of the stain condition on the surface of the work target. It provides key data support for the cleaning operation of the execution control module and the abnormal tracing of the analysis and feedback module, ensuring that the cleaning system can formulate reasonable cleaning strategies based on the actual stain situation and improve the efficiency and effectiveness of the cleaning operation.

[0093] Example 3:

[0094] The task definition module in this embodiment is implemented as follows: The core function of this module is to obtain the physical parameters of the target surface and define preset cleaning indicators based on these parameters, providing standards and a basis for subsequent cleaning operations. The specific working process is to first obtain the physical parameters of the target surface. The target surface can be different components of the drone, such as the fuselage skin, wing surface, propeller blades, etc. Different components have different materials and structures, and their physical parameters also vary. Therefore, when obtaining the physical parameters, it is necessary to select appropriate measurement methods and instruments based on the specific conditions of the target surface.

[0095] Surface roughness is one of the important physical parameters of the target surface, which reflects the microscopic geometric characteristics of the surface. Surface roughness can be measured by contact measurement or non-contact measurement. The contact measurement method usually uses a roughness meter to record the fluctuations of the surface profile through the contact between the probe and the surface, thereby obtaining surface roughness parameters such as Ra (arithmetic mean deviation of the profile) and Rz (maximum height of the profile). Non-contact measurement methods can use optical technologies such as laser interferometry and light sectioning to obtain surface roughness information by analyzing the reflected light from the surface. For certain parts of drones, such as wing surfaces, higher surface roughness may affect aerodynamic performance, as well as the adhesion of stains and the difficulty of cleaning. Therefore, accurate measurement of surface roughness is crucial for formulating cleaning strategies.

[0096] Stain adhesion refers to the strength of the bond between a stain and the target surface, which directly affects the ease of cleaning. There are various methods for measuring stain adhesion, including stretching, peeling, and shearing. Taking the stretching method as an example, a certain amount of stain is applied to the target surface. After it dries, a tensile testing machine is used to apply a tensile force perpendicular to the surface. The maximum tensile force at which the stain is peeled off is recorded, which reflects the strength of the stain's adhesion. Different types of stains, such as oily, water-based, and granular, have significantly different adhesion strengths. The material and roughness of the target surface also affect stain adhesion. Therefore, measurements must be tailored to the specific stain type and target surface to obtain accurate stain adhesion data.

[0097] The maximum water pressure limit refers to the maximum water pressure that can be applied without damaging the target surface. Determining the maximum water pressure limit requires considering factors such as the material strength, structural characteristics, and connection method of the target surface. UAV components, such as fuselage skins, are typically made of composite materials or aluminum alloys. These materials have a certain compressive strength, and exceeding this strength may cause surface deformation, cracking, or coating damage. The maximum water pressure limit can be determined by calculating material mechanics and combining it with the structural design of the target surface to estimate the maximum water pressure it can withstand. Alternatively, the maximum water pressure limit can be determined through experimental testing, where water pressure is gradually applied to the target surface to observe for signs of damage. During the cleaning process, if the water pressure exceeds this limit, irreversible damage to the drone components may occur. Therefore, accurately obtaining the maximum water pressure limit is an important prerequisite for ensuring the safe conduct of cleaning operations.

[0098] The chemical activity threshold of a detergent refers to the maximum chemical activity index that the detergent can achieve without causing adverse reactions with the target surface. If the chemical activity of the detergent is too high, it may cause chemical reactions such as corrosion and dissolution with the material of the target surface, resulting in surface damage; while if the chemical activity is too low, it may not be able to effectively remove stains. The chemical activity threshold of a detergent is usually related to factors such as the composition, concentration, and pH value of the detergent. The method for determining the chemical activity threshold of a detergent can be to conduct chemical analysis and material compatibility testing, and conduct contact experiments between detergents of different chemical activities and the target surface material to observe whether the material shows abnormal phenomena such as discoloration, swelling, and strength loss, thereby determining the chemical activity threshold of the detergent. For example, for aluminum alloy surfaces, if the pH value of the detergent is too high, it may cause corrosion of the aluminum alloy, so it is necessary to determine its allowable pH value range, that is, the chemical activity threshold.

[0099] After obtaining the physical parameters of the target surface, preset cleaning indicators are defined based on these parameters. The preset cleaning indicators include a preset stain density gradient threshold, a preset stain point missing rate threshold, a preset water pressure adjustment threshold, etc.

[0100] The preset stain density gradient threshold is used to determine whether the difference in stain density on the target surface is within an acceptable range. The setting of this threshold needs to comprehensively consider the cleaning requirements of the target surface and the difficulty of stain removal. For example, for the optical lens surface of a drone, the cleaning accuracy requirements are high, and the preset stain density gradient threshold should be set low to ensure that the stains on the lens surface are thoroughly removed; while for the fuselage surface, the cleaning accuracy requirements are relatively low, and the preset stain density gradient threshold can be appropriately increased. When setting the preset stain density gradient threshold, you can refer to the stain density data of the benchmark model in the stain distribution modeling module, and combine it with the cleaning standard of the target surface to determine a reasonable threshold range. When the stain density gradient on the target surface exceeds the threshold, it indicates that more effective cleaning measures are needed.

[0101] The preset stain point missing rate threshold is used to measure the integrity and effectiveness of the inspection points. As mentioned above, due to various reasons, the number of inspection points may not match the number of reference points. If the stain point missing rate is too high, it will affect the accurate judgment of the stain condition on the surface of the work target. The setting of the preset stain point missing rate threshold needs to consider factors such as the performance of the inspection equipment and the complexity of the work target. For work targets with simple shapes and good inspection conditions, the preset stain point missing rate threshold can be set lower; for work targets with complex shapes and high inspection difficulty, the preset stain point missing rate threshold can be appropriately increased, but it is necessary to ensure that the missing rate is within a reasonable range to ensure the reliability of the inspection data.

[0102] The preset water pressure adjustment threshold is an important basis for dynamic response water pressure adjustment during cleaning. The setting of this threshold is based on the maximum water pressure limit, while taking into account the cleaning effect and safety. The preset water pressure adjustment threshold is usually lower than the maximum water pressure limit to leave a certain safety margin. For example, if the maximum water pressure limit is 10MPa, the preset water pressure adjustment threshold can be set to 8MPa. This ensures the cleaning effect while avoiding exceeding the maximum water pressure limit due to water pressure adjustment errors, which may cause damage to the target surface. The preset water pressure adjustment threshold also needs to be adjusted according to the type and adhesion of the stain. For stains with high adhesion, the preset water pressure adjustment threshold can be appropriately increased to enhance the cleaning effect; for stains with low adhesion, the preset water pressure adjustment threshold can be lowered to reduce energy consumption and the risk of damage to the surface.

[0103] Example 4:

[0104] The implementation of the execution control module in this embodiment is as follows: The core function of this module is to perform steady-state cleaning, dynamic response cleaning, and composite cleaning operations on the work target located in the path planning framework. Through a series of calculations, judgments, and adjustments, the effective cleaning of the work target is achieved. The specific working process is as follows: First, steady-state cleaning is performed on the work target located in the path planning framework. In the steady-state cleaning stage, the coordinates and number of the detection points and reference points need to be obtained. These coordinate data come from the detection point coordinate set recorded in the sensor configuration module. and the reference point coordinate set recorded in the stain distribution modeling module .

[0105] Calculate the stain density gradient based on the acquired coordinate data and stain point missing rate . Stain density gradient The calculation method is to take the difference in stain density between the detection point and the corresponding reference point to reflect the difference in stain distribution between the target surface and the reference model; the stain point missing rate The calculation method is and The ratio of is the number of benchmark points, is the number of test points, which is used to measure the difference between the number of test points and the number of benchmark points.

[0106] The calculated stain density gradient and stain point missing rate Compare it with the preset stain density gradient threshold and the preset stain point missing rate threshold defined in the task definition module. and If the corresponding threshold requirements are met at the same time, it means that the stain condition on the surface of the work target is within an acceptable range, and the work target is judged to have passed steady-state cleaning; if not, it is necessary to adjust the cleaning parameters or take other cleaning measures.

[0107] For the operation goal of steady-state cleaning, a dynamic response cleaning environment is constructed. In the dynamic response cleaning environment, water pressure adjustment, detergent ratio adjustment and spray uniformity adjustment operations are performed. Dynamic response factor Including water pressure response factor and detergent response factors , where the water pressure response factor It is divided into high adhesion area water pressure factor and low adhesion area water pressure factor, which are used to adjust the water pressure in high adhesion area and low adhesion area respectively; spray uniformity factor It includes axial uniformity factor and radial uniformity factor, which are used to measure and adjust the uniformity of spraying in the axial and radial directions of the operating target.

[0108] When building a dynamic response cleaning environment, step-by-step water pressure and detergent concentration commands are applied to the target. Step-by-step water pressure commands gradually adjust the water pressure at specific intervals and pressure gradients, for example, starting at a lower pressure and gradually increasing it to the set target pressure to accommodate the cleaning needs of different soiled areas. Step-by-step detergent concentration commands adjust the detergent concentration in stages, using different concentrations for different soil types and levels of soil adhesion.

[0109] The adjustment results are output in the form of a quantitative matrix, which contains various adjustment data such as water pressure adjustment parameters, detergent ratio parameters, spray uniformity parameters, etc., which is convenient for intuitive understanding and evaluation of the adjustment effect.

[0110] During the dynamic response cleaning process, the adjustment effect needs to be tested and judged. The water pressure response test result is the water pressure rate in the high adhesion area. and water pressure rate in the low adhesion area ,in Indicates how quickly the water pressure changes in areas with high soil adhesion. Indicates how fast the water pressure changes in the low-adhesion stain area; the spray uniformity test result is the axial deviation and radial deviation , Used to measure the uniformity deviation of the spray in the axial direction of the working target. Used to measure the uniformity deviation of the spray in the radial direction.

[0111] The water pressure rate in the high adhesion area , water pressure rate in low adhesion area , axial deviation and radial deviation Compare this to the preset water pressure adjustment threshold and spray uniformity threshold defined in the task definition module. If these test results meet the corresponding threshold requirements, dynamic response cleaning has achieved the expected results and the job objective has been determined to have passed dynamic response cleaning. If not, further optimization of the adjustment parameters is required.

[0112] For dynamic response cleaning targets, complex cleaning operations are performed to simulate cleaning response behavior under varying operating conditions. These varying operating conditions include factors such as flight speed, ambient temperature, and humidity, which can affect cleaning effectiveness. Therefore, the stability and effectiveness of the cleaning system under these varying conditions must be verified.

[0113] During the composite cleaning process, the comprehensive cleaning stability coefficient λ of the operation target is calculated, and the calculation formula is:

[0114]

[0115] in: It is a comprehensive cleaning stability coefficient, used to evaluate the cleaning stability of the operating target under variable working conditions; is the number of variable operating conditions, that is, the number of different operating conditions simulated; and is the weight coefficient used to measure the dynamic response factor and spray uniformity factor The importance of the comprehensive stability assessment can be set according to the actual cleaning needs and working conditions; Indicates the The dynamic response factor under a variable working condition reflects the response of water pressure and detergent under the working condition; Indicates the The spray uniformity factor under a variable working condition reflects the spray uniformity under the working condition.

[0116] By calculating the comprehensive cleaning stability coefficient λ, the working performance of the cleaning system under different variable working conditions can be comprehensively evaluated, providing a basis for further optimizing the cleaning strategy and improving the cleaning effect.

[0117] The implementation process of the entire execution control module realizes multi-stage and refined cleaning of the operation target through the sequential execution of steady-state cleaning, dynamic response cleaning and compound cleaning, combined with the calculation, comparison and adjustment of various parameters, ensuring that the cleaning system can adapt to different stain conditions and working conditions to achieve efficient and reliable cleaning effects.

[0118] Example 5:

[0119] The implementation of the analysis and feedback module in this embodiment is as follows: The core function of this module is to trace the abnormality of the operation target that fails the dynamic response cleaning, generate diagnostic logs and update the cleaning strategy, and realize data storage and transmission through the data aggregation module and communication relay module. The specific working process is that when the operation target fails the dynamic response cleaning, the cleaning data of the operation target is obtained, including the detection point coordinate set. , reference point coordinate set , stain density gradient , dynamic response factor And other data.

[0120] Calculate the stain density gradient direction vector based on the acquired cleaning data ,in is the coordinate vector of the reference point, is the detection point coordinate vector, The calculation method is the difference between the detection point coordinate vector and the reference point coordinate vector, that is: Stain density gradient direction vector It can reflect the changing direction and trend of the stain density on the target surface, and provide a basis for analyzing the conduction path of the stain residue.

[0121] Based on the calculated stain density gradient direction vector , analyzing the conduction path of residual stains. This analysis can locate blind spots or areas with high residue. Blind spots typically refer to areas where the cleaning fluid cannot effectively reach due to factors such as the drone's flight path, spray angle, or water pressure. High residue areas are areas with a large gradient of stain density and a high concentration of residual stains. These areas require special attention and treatment during subsequent cleaning operations.

[0122] The target is divided into a finite number of equal-thickness segments along the axial direction. The choice of axial direction can be determined based on the target's structural characteristics and the coordinate system of the path planning framework, such as segmentation along the path planning framework's z-axis. A preset inspection point density threshold is defined, based on the target's cleaning requirements and inspection accuracy requirements.

[0123] The number of inspection points within each section is counted, and the section inspection point density is calculated as the ratio of the number of inspection points within the section to the section's volume or area. Sections that exceed the preset inspection point density threshold undergo surface roughness verification. A density exceeding the threshold may indicate complex surface conditions, such as numerous uneven surfaces or gaps, which can easily lead to stains remaining. Therefore, further surface roughness verification is necessary to determine whether adjustments to the cleaning strategy are necessary, such as increasing water pressure, adjusting the spray angle, or changing the detergent.

[0124] In addition, the system also includes a data aggregation module and a communication relay module. The data aggregation module is connected with the stain distribution modeling module and the sensor configuration module to periodically obtain the reference point coordinate set. and the detection point coordinate set , construct the stain gradient distribution matrix. The data aggregation cycle is synchronized with the execution cycle of dynamic response cleaning to ensure that the acquired data can timely reflect the stain changes on the target surface. The dimension of the stain gradient distribution matrix and the number of reference points Each element in the matrix corresponds to the stain gradient value of a reference point or detection point, which facilitates the visualization and analysis of the stain gradient distribution and stores it in the local database to provide support for subsequent cleaning strategy optimization and historical data query.

[0125] The communication relay module connects to the external flight control system and encapsulates the control module's adjustment results and the analysis and feedback module's diagnostic log into flight control protocol data packets. These adjustment results include parameters such as water pressure adjustment, detergent ratio, and spray uniformity. The diagnostic log contains analysis results for abnormality tracing, location information for cleaning blind spots and high-residue areas, and surface roughness verification results. The flight control protocol data packets are transmitted to the flight control system via a wireless network, enabling collaboration with the drone's flight control system. This allows the flight control system to adjust the drone's flight path, speed, and other parameters based on feedback from the cleaning system, ensuring smooth cleaning operations and optimal cleaning results.

[0126] The implementation process of the entire analysis and feedback module achieves accurate tracing and diagnosis of cleaning anomalies through data calculation, path analysis, section detection and density verification of work targets that fail dynamic response cleaning. Combined with the data storage and transmission functions of the data aggregation module and the communication relay module, it provides strong support for the update of cleaning strategies and the adjustment of UAV flight control, enabling the cleaning system to be continuously optimized and improved, thereby improving cleaning efficiency and reliability.

[0127] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0128] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A cleaning application and data processing system suitable for drones, characterized in that: include: Stain distribution modeling module: Scans the target cleaning surface structure to generate a stain distribution model, constructs a spatial coordinate system with the geometric center of the stain distribution model, uses the stain distribution model as a reference model, and uses the spatial coordinate system as a path planning framework to create reference points of the reference model at preset intervals. The reference points include basic stain points, key stain points, and matching stain points. The basic stain points and reference points evenly cover the stain distribution model, and the spacing between adjacent reference points is equal. The key stain points are the reference points of the high-adhesion stain area, and the matching stain points are the reference points of the axial trisection section of the operation target. Sensor configuration module: The surface to be cleaned is defined as the work target, which is placed in the path planning framework. Inspection points for the work target are arranged at preset intervals, including basic inspection points, key inspection points, and matching inspection points. Basic inspection points evenly cover the work target, and adjacent inspection points are spaced equally. Key inspection points are inspection points in areas with high stain adhesion on the work target, and matching inspection points are inspection points that divide the work target into three equal sections along the axial direction. The benchmark model is aligned with the work target, and the difference in stain density between the inspection points and the benchmark points is defined as the stain density gradient. Task definition module: obtains the physical parameters of the target surface and defines preset cleaning indicators based on the physical parameters; Execution control module: Executes cleaning operations on the job target, including steady-state cleaning, dynamic response cleaning, and composite cleaning. Steady-state cleaning determines whether it passes by judging the stain density gradient and stain point loss rate. Dynamic response cleaning includes water pressure, detergent ratio, and spray angle adjustment. Composite cleaning simulates the cleaning response behavior under variable working conditions. Analysis and feedback module: Trace the anomalies of job targets that fail dynamic response cleaning, generate diagnostic logs and update cleaning strategies.

2. The cleaning application and data processing system for drones according to claim 1, characterized in that: The stain distribution modeling module includes: Scan the target cleaning surface through multispectral imaging to build a stain distribution model; Establish a spatial coordinate system based on the geometric center of the stain distribution model, generate reference points including basic stain points, key stain points and matching stain points, and record the coordinate set of the reference points , is the number of benchmark points; The stain distribution model is used as the benchmark model and the spatial coordinate system is used as the path planning framework.

3. The cleaning application and data processing system for drones according to claim 1, characterized in that: The sensor configuration module includes: Define the surface to be cleaned as the operation target, locate the operation target in the path planning framework, lay out the detection points including basic detection points, key detection points and matching detection points, and record the detection point coordinate set , is the number of detection points; Overlapping the matching benchmark points with the matching detection points to achieve alignment between the work target and the benchmark model; The difference in stain density between the detection point and the reference point is defined as the stain density gradient , including the base point gradient , key point gradient , matching point gradient ; Get the number of benchmark points and the number of testing points ,definition and The ratio is the stain point missing rate .

4. The cleaning application and data processing system for drones according to claim 1, characterized in that: The task definition module includes: Obtaining physical parameters of the target surface including surface roughness, stain adhesion, maximum water pressure limit, and cleaning agent chemical activity threshold; The preset cleaning indicators are defined according to the physical parameters, including the preset stain density gradient threshold, the preset stain point missing rate threshold, and the preset water pressure adjustment threshold.

5. The cleaning application and data processing system for drones according to claim 3, characterized in that: The execution control module includes: Perform steady-state cleaning, dynamic response cleaning, and composite cleaning on the work targets located in the path planning framework; The steps of steady-state cleaning are: obtaining the coordinates and number of detection points and reference points, calculating the stain density gradient and stain point missing rate ,like and If the preset dirt density gradient threshold and the preset dirt point missing rate threshold are met at the same time, the operation target is determined to have passed the steady-state cleaning; A dynamic response cleaning environment is constructed for the steady-state cleaning operation target, and water pressure, detergent ratio and spray uniformity are adjusted. The adjustment results are output in a quantitative matrix. Composite cleaning is performed on the job targets through dynamic response cleaning to simulate the cleaning response behavior under variable working conditions.

6. The cleaning application and data processing system for drones according to claim 5, characterized in that: The steps to build a dynamic response cleaning environment and output the quantization matrix are: Define dynamic response factors for job objectives Including water pressure response factor , detergent response factor , spray uniformity factor ; The water pressure response factor Including high adhesion area water pressure factor and low adhesion area water pressure factor, spray uniformity factor Including axial uniformity factor and radial uniformity factor; Build a dynamic response cleaning environment, including step-by-step water pressure instructions and step-by-step detergent concentration instructions applied to the work target; The water pressure response test result is defined as the water pressure rate in the high adhesion area and water pressure rate in the low adhesion area , the spray uniformity test result is axial deviation and radial deviation ; like 、 、 、 If the corresponding preset water pressure regulation threshold and the preset spray uniformity threshold are met, it is determined that the operation target has passed the dynamic response cleaning.

7. The cleaning application and data processing system for drones according to claim 5, characterized in that: The steps of performing composite cleaning are: The comprehensive cleaning stability coefficient of the calculated operation target is: in, is the comprehensive cleaning stability coefficient, is the number of variable working condition scenarios, and is the weight coefficient, Indicates the The dynamic response factor under a variable working condition scenario, Indicates the The spray uniformity factor under different working conditions.

8. The cleaning application and data processing system for drones according to claim 1, characterized in that: The analysis feedback module includes: Obtain cleaning data for targets that fail dynamic response cleaning and calculate the direction vector of the stain density gradient ,in is the coordinate vector of the reference point, is the detection point coordinate vector; based on Analyze the conduction path of residual stains and locate cleaning blind spots or high residue areas; The work target is divided into a finite number of equal-thickness sections along the axial direction, a preset detection point density threshold is defined, the number of detection points in each section is counted, the section detection point density is calculated, and the surface roughness verification is performed on sections that exceed the preset threshold.

9. The cleaning application and data processing system for drones according to claim 3, characterized in that: The system further comprises: The data aggregation module is connected with the stain distribution modeling module and the sensor configuration module to periodically obtain the reference point coordinate set and the detection point coordinate set , construct the stain gradient distribution matrix, store the stain gradient distribution matrix in the local database, where the data aggregation cycle is synchronized with the execution cycle of dynamic response cleaning, the dimension of the stain gradient distribution matrix and the number of reference points consistent.

10. The cleaning application and data processing system for drones according to claim 5, characterized in that: The system further comprises: The communication relay module is connected to the external flight control system, encapsulates the adjustment results of the execution control module and the diagnostic log of the analysis and feedback module into a flight control protocol data packet, and transmits it to the flight control system through a wireless network.

Images