An unmanned aerial vehicle-based outer wall cleaning method and unmanned aerial vehicle
By acquiring images of the glass surface and the bottom of the frame, analyzing the level of contamination and dust, and adjusting the suction power and spray flow of the dust collection component, the problem of dust from the bottom of the frame contaminating the glass below during drone cleaning was solved, achieving a more efficient cleaning effect.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XINGWEI LINGTUO (SHANXI) TECH CO LTD
- Filing Date
- 2026-03-13
- Publication Date
- 2026-06-05
Smart Images
Figure CN122140157A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of drones, and in particular to a drone-based method for cleaning exterior walls and the drone itself. Background Technology
[0002] With the development of the low-altitude economy and drone technology, drones are being used in an increasing number of scenarios, such as cleaning the glass facades of high-rise buildings. Currently, drone cleaning of glass facades mainly involves attaching a nozzle to the drone to extract cleaning agent, spraying it onto the glass facade, and then using the nozzle to spray high-pressure water onto the glass to rinse it, thus achieving a cleaning effect. However, this method can only clean dust and other surface dirt on the glass facade. Since high-rise buildings consist of numerous glass units, each including glass and a frame that holds the glass in place, a large amount of dust accumulates at the bottom of the frame. When cleaning the glass using the above method, the mixture of dust and water flows down the glass frame, contaminating the already cleaned glass below. Therefore, how to reduce the contamination of the glass below by dust deposited at the bottom of the glass frame during drone cleaning of glass facades has become a problem. Summary of the Invention
[0003] To reduce the contamination of the glass below by dust deposited at the bottom of the glass frame when using drones to clean exterior walls, this application provides a drone-based exterior wall cleaning method and a drone.
[0004] Firstly, this application provides a method for cleaning exterior walls based on drones, employing the following technical solution: A method for cleaning exterior walls using drones includes: Acquire images of the glass surface of the glass unit on the outer wall at the location of the drone, as well as the bottom image of the glass unit's frame; The degree of contamination on the glass surface is determined based on the glass surface image, and the degree of dust deposition at the bottom of the frame is determined based on the bottom image. The suction power of the dust collection component is determined based on the dust deposition level, and the operation of the dust collection component is controlled based on the suction power to extract the dust from the bottom of the frame. The spray flow rate and moving speed of the drone when spraying cleaning agent onto the glass surface are determined based on the pollution level value and the dust deposition level value. The drone is controlled to operate based on the stated movement speed and to spray cleaning agent according to the stated spray flow rate.
[0005] By employing the above technical solution, images of the glass surface and bottom are obtained. These images record the specific contamination status of the glass surface and bottom edge. Therefore, the degree of contamination on the glass surface can be accurately determined based on the surface image. Similarly, the degree of dust deposition at the bottom of the edge can be accurately determined based on the bottom image. A higher dust deposition value indicates more dust accumulation, which in turn indicates a greater likelihood that the dust-water mixture will flow along the edge to the lower glass surface during subsequent rinsing. Therefore, a stronger suction is needed to remove the dust from the bottom of the edge. The appropriate suction power of the vacuuming component is determined, and its operation is controlled according to this value to better remove dust from the bottom of the frame. Based on a comprehensive analysis of the degree of contamination and dust deposition on the glass surface, the spray flow rate and the drone's movement speed are determined when applying cleaning agent. This ensures that the cleaning agent is sprayed more thoroughly and comprehensively onto the glass frame, resulting in more complete dissolution of stains and mixing of dust. During subsequent rinsing, the mixture of dust and water is less likely to flow to the lower glass surface, keeping the lower glass surface clean and reducing contamination of the cleaned glass below.
[0006] In another possible implementation, determining the degree of contamination of the glass surface based on the glass surface image includes: Feature recognition is performed on the glass surface image to obtain the dirt features on the glass surface; Determine the area and gray value of each dirt feature, and count all dirt features to obtain the first quantity of all dirt. The glass surface image is subjected to grayscale transformation to obtain a surface grayscale image, and the average grayscale value of the surface grayscale image is determined. Determine the absolute value of the difference between the average gray value and the preset gray value; The degree of contamination on the glass surface is determined based on the area, gray value, first quantity, and absolute value of the difference of each dirt feature.
[0007] In another possible implementation, determining the degree of contamination of the glass surface based on the area, grayscale value, first quantity, and absolute value of the difference of each dirt feature includes: The sub-contamination value of each contamination feature is determined based on the area and gray value of each contamination feature; The total contamination value for all contamination features is determined based on the sub-contamination value of each contamination feature. The degree of contamination on the glass surface is determined based on the total contamination value, the first quantity, and the absolute value of the difference.
[0008] In another possible implementation, determining the spray flow rate and drone speed when spraying cleaning agent onto the glass surface based on the contamination level value and the dust deposition level value includes: The score of the glass unit is determined based on the pollution level value, the dust deposition level value, and their respective weights; The target score interval is determined from multiple preset score intervals. Each preset score interval corresponds to a preset spray flow rate and a preset moving speed. The preset spray flow rate and preset moving speed of the target score interval are determined as the spray flow rate and moving speed of the drone when spraying cleaning agent onto the glass surface.
[0009] In another possible implementation, the bottom image includes a first image of the horizontal plane at the bottom of the border and a second image of the vertical plane of the outer wall at the bottom of the border. Determining the degree of dust deposition at the bottom of the border based on the bottom image includes: Perform a grayscale transformation on the first image to obtain a first grayscale image; Determine the first grayscale value histogram of the first grayscale image, and calculate the first similarity between the first grayscale value histogram and the first preset histogram, wherein the first preset histogram is the grayscale value histogram when there is no dust deposition on the bottom horizontal plane of the border; The second image is subjected to grayscale transformation to obtain a second grayscale image; Determine the second grayscale value histogram of the second grayscale image, and calculate the second similarity between the second grayscale value histogram and the second preset histogram, wherein the second preset histogram is the grayscale value histogram when there is no dust deposition on the vertical surface of the bottom outer wall of the frame; The degree of dust deposition at the bottom of the border is determined based on the first similarity and the second similarity.
[0010] In another possible implementation, the method further includes: When the drone is detected to have reached the bottom edge of the glass unit, the spray flow rate is increased.
[0011] In another possible implementation, determining the suction power of the dust collection component based on the dust deposition level includes: A preset function for calculating the suction value is retrieved, and the dust deposition level value is substituted into the preset function to obtain the suction value.
[0012] Secondly, this application provides a drone, which adopts the following technical solution: A drone, comprising: The drone itself; A camera device is mounted on the drone body to capture images of the glass surface of the glass unit on the outer wall where the drone is located, as well as the bottom image of the glass unit's frame. A dust-collecting component, mounted on the drone body, is used to collect dust from the bottom of the frame. Spraying assembly for spraying cleaning agents; At least one processor, wherein the camera device, the dust collection assembly, and the spraying assembly are all communicatively connected to the at least one processor; Memory; At least one application, wherein the at least one application is stored in the memory and configured to be executed by the at least one processor, the at least one application: executing a drone-based exterior wall cleaning method according to any possible implementation of the first aspect.
[0013] By adopting the above technical solution, the camera device facilitates the acquisition of images of the glass surface and bottom, the dust suction component facilitates the removal of dust from the bottom of the frame, and the spraying component facilitates the spraying of cleaning agent onto the glass unit. These images record the specific contamination levels on the glass surface and bottom frame. Therefore, the degree of contamination on the glass surface can be accurately determined based on the surface image. Similarly, the degree of dust deposition at the bottom of the frame can be accurately determined based on the bottom image. A higher dust deposition value indicates more dust accumulation, which in turn indicates a greater likelihood that the dust-water mixture will flow along the frame to the lower glass surface during subsequent rinsing. This requires greater suction to remove dust from the bottom of the frame. The appropriate suction power of the vacuuming component is determined based on the degree of dust deposition, and the vacuuming component is controlled according to the determined suction power to better remove dust from the bottom of the frame. The spray flow rate and drone speed when spraying cleaning agent are determined by comprehensively analyzing the degree of contamination and dust deposition on the glass surface. This allows the cleaning agent to be sprayed more thoroughly and comprehensively onto the glass frame, making the stains dissolve and the dust mix more completely. During subsequent rinsing, the mixture of dust and water is less likely to flow to the lower glass surface, keeping the lower glass surface clean and reducing contamination of the cleaned glass below.
[0014] In another possible implementation, the dust collection assembly includes a motor mounted on the drone body, a gear mounted on the motor, a support frame mounted on the drone body, a rack mounted within the support frame, a dust collection pipe mounted on the rack, and a blower communicating with the dust collection pipe.
[0015] By adopting the above technical solution, the motor drives the gear to rotate, the gear to rotate and the rack to move, and the rack to move and the suction pipe to move. This allows the suction pipe to move and suck up dust from different positions on the bottom edge of the drone without the drone needing to make lateral horizontal displacement.
[0016] In another possible implementation, a notch is provided at the port of the suction pipe, and the notch is located below the suction pipe.
[0017] By adopting the above technical solution, after the notch is opened, the suction pipe at the notch contacts the horizontal and vertical surfaces at the bottom of the frame respectively, thereby simultaneously sucking up the dust deposited on the horizontal surface and the dust attached to the vertical surface, thus improving the suction efficiency.
[0018] Thirdly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium that, when the computer program is executed in a computer, causes the computer to perform a drone-based exterior wall cleaning method as described in any of the first aspects.
[0019] In summary, this application includes at least one of the following beneficial technical effects: The process involves acquiring images of the glass surface and bottom edge, which record the specific level of contamination on these areas. The surface image accurately determines the degree of contamination, while the bottom image accurately determines the degree of dust accumulation at the bottom edge. A higher dust accumulation value indicates more dust buildup, meaning a greater likelihood of the dust-water mixture flowing down the edge to the lower glass surface during subsequent rinsing. Therefore, a stronger suction is needed to remove the dust from the bottom edge. The appropriate suction power for the vacuum assembly is determined based on the dust accumulation value, and the assembly is controlled accordingly to better remove dust. A comprehensive analysis of the surface contamination and dust accumulation values determines the spray flow rate and drone speed for applying cleaning agent. This ensures the cleaning agent is applied more thoroughly to the glass edge, resulting in more complete dissolution and dust mixing. This reduces the likelihood of the dust-water mixture flowing down to the lower glass surface during subsequent rinsing, keeping the lower glass surface clean and minimizing contamination of the cleaned glass. Attached Figure Description
[0020] Figure 1 This is a flowchart illustrating an embodiment of an exterior wall cleaning method based on a drone.
[0021] Figure 2 This is a schematic diagram of the structure of a drone according to an embodiment of this application.
[0022] Figure 3 This is another structural schematic diagram of a drone according to an embodiment of this application.
[0023] Reference numerals: 1. UAV body; 2. Camera device; 3. Dust suction assembly; 31. Motor; 32. Gear; 33. Support frame; 34. Rack; 35. Dust suction pipe; 351. Notch; 36. Blower; 4. Spraying assembly; 41. Spray nozzle; 42. Water pump; 51. Processor; 52. Bus; 53. Memory; 54. Transceiver. Detailed Implementation
[0024] The present application will be further described in detail below with reference to the accompanying drawings.
[0025] After reading this specification, those skilled in the art may make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they fall within the scope of the claims of this application.
[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0027] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.
[0028] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.
[0029] This application provides a method for cleaning exterior walls using a drone, which is executed by a drone, such as... Figure 1 As shown, the method includes steps S101, S102, S103, S104, and S105, wherein, S101, acquire images of the glass surface of the glass unit on the outer wall at the location of the drone and the bottom image of the glass unit's frame.
[0030] In this embodiment, the drone can fly vertically downwards from the rooftop of a high-rise building. During flight, the drone's onboard camera captures images of the glass surface and the bottom of the frame of each glass unit. The exterior glass curtain wall typically includes frames for fixing windows and glass within those frames, and the glass facade of a high-rise building consists of numerous glass units. When the drone reaches a new glass unit, it can pause briefly. Based on the acquired surface and bottom images, the drone assesses the degree of contamination of the glass, thereby determining the appropriate cleaning agent spray flow rate and the drone's downward flight speed for spraying the cleaning agent. After a brief hover, it operates according to the determined spray flow rate and speed.
[0031] S102, determine the degree of contamination on the glass surface based on the glass surface image, and determine the degree of dust deposition at the bottom of the frame based on the bottom image.
[0032] In this embodiment of the application, since the glass surface image records the degree of contamination on the glass surface, the drone can accurately determine the degree of contamination by analyzing the glass surface image. Similarly, the bottom image records the dust deposition at the bottom of the frame, so the drone can accurately determine the degree of dust deposition at the bottom of the frame by analyzing the bottom image.
[0033] S103, determine the suction value of the dust collection component based on the dust deposition level value, and control the operation of the dust collection component based on the suction value to suck up the dust at the bottom of the frame.
[0034] In this embodiment, the more dust accumulates at the bottom of the frame, the more the dust mixes with water and flows down to the clean glass surface when the glass unit is cleaned later, causing re-contamination of the clean glass surface. Therefore, the drone is equipped with a dust collection component. The drone first determines the appropriate suction value based on the degree of dust accumulation. For example, the more dust accumulates, the greater the required suction power. After determining the appropriate suction value, the drone controls the dust collection component to operate according to the determined suction value, thereby sucking away the dust at the bottom of the frame and reducing the accumulation of dust at the bottom of the frame. When cleaning with water later, the amount of dust mixed in the water is reduced, thereby reducing the re-contamination of the glass below and making the glass less likely to get dirty again.
[0035] S104, Based on the pollution level value and the dust deposition level value, determine the spray flow rate and the drone's movement speed when spraying cleaning agent onto the glass surface.
[0036] In the embodiments of this application, the higher the pollution level value and the dust deposition level value, the dirtier and more serious the pollution on the glass surface, requiring more cleaning agent to be sprayed, and requiring more comprehensive and detailed spraying of the glass unit. Therefore, the drone comprehensively considers and analyzes the pollution level value and the dust deposition level value to obtain the appropriate spraying flow rate and the drone's moving speed when spraying cleaning agent. The higher the spraying flow rate, the more cleaning agent is sprayed per unit time, and the greater the impact force of the cleaning agent.
[0037] S105 controls the operation of the drone based on its movement speed and sprays cleaning agent according to the spray flow rate.
[0038] In this embodiment, after determining the appropriate spray flow rate and moving speed, the drone begins spraying cleaning agent onto the glass unit according to the spray flow rate and moving speed. This allows the cleaning agent to better mix with stains and dust on the glass unit, facilitating subsequent rinsing and cleaning. Since dust at the bottom of the frame has already been absorbed, the water flowing downwards during subsequent rinsing is cleaner, reducing the risk of re-contamination of the already cleaned glass below.
[0039] One possible implementation of this application embodiment is that step S102, which determines the degree of contamination of the glass surface based on the glass surface image, specifically includes steps S1021 (not shown in the figure), S1022 (not shown in the figure), S1023 (not shown in the figure), S1024 (not shown in the figure), and S1025 (not shown in the figure), wherein... S1021, Perform feature recognition on the glass surface image to obtain the dirt features on the glass surface.
[0040] In this embodiment, the drone can perform edge detection on a glass surface image. First, the glass surface image is denoised. Then, a grayscale transformation is performed on the denoised image to obtain a grayscale image. The locations where grayscale values jump in the grayscale image represent dirt features. Alternatively, the drone can input the glass surface image into a trained network model for feature recognition to obtain dirt features on the glass surface. The network model can be a convolutional neural network model or other types of network models capable of feature recognition. Dirt features include bird droppings, insect carcasses, and plant secretions, etc.
[0041] S1022, determine the area and gray value of each dirt feature, and count all dirt features to obtain the first quantity of all dirt.
[0042] In this embodiment, a larger area of dirt features indicates a greater degree of contamination on the glass surface. The drone can count the number of pixels within the dirt feature area, using the pixel count to represent the area of each dirt feature. The drone performs grayscale transformation on the glass surface image of the area containing the dirt feature to obtain a grayscale image of that area. Then, it averages the grayscale values within that area to obtain a grayscale value representing the color depth of each dirt feature. A larger grayscale value indicates a more stubborn dirt feature, which more significantly affects the degree of contamination on the glass surface. The drone counts all dirt features on the glass surface to obtain a first quantity of all dirt. A larger first quantity indicates a higher degree of contamination on the glass surface.
[0043] S1023, perform grayscale transformation on the glass surface image to obtain a surface grayscale image, and determine the average grayscale value of the surface grayscale image.
[0044] In this embodiment of the application, the UAV performs noise reduction processing on the glass surface image, then performs grayscale transformation on the noise-reduced image to obtain a surface grayscale image, and then the UAV calculates the average grayscale value of the surface grayscale image using the average value calculation formula to obtain the average grayscale value.
[0045] S1024, Determine the absolute value of the difference between the average gray value and the preset gray value.
[0046] In this embodiment of the application, a preset gray value is used as the gray value when the glass unit is clean. The drone subtracts the preset gray value from the average gray value to obtain the difference, and then takes the absolute value of the difference. The larger the absolute value of the difference, the higher the degree of contamination on the current glass surface.
[0047] S1025, determine the degree of contamination of the glass surface based on the area, gray value, first quantity, and absolute value of the difference of each dirt feature.
[0048] In summary, for the embodiments of this application, the area, gray value, first number of all dirt features, and absolute value of the difference of each dirt feature are all key factors affecting the degree of contamination of the glass surface. Therefore, the UAV can accurately determine the degree of contamination of the glass surface by comprehensively analyzing the above four factors.
[0049] One possible implementation of this application embodiment is that step S1025 determines the degree of contamination on the glass surface based on the area, grayscale value, first quantity, and absolute value of the difference of each dirt feature. Specifically, this includes steps Sa (not shown in the figure), Sb (not shown in the figure), and Sc (not shown in the figure). Sa determines the sub-fouling value for each fouling feature based on its area and grayscale value.
[0050] In the embodiments of this application, the area and gray value of each dirt feature are factors that affect the degree of contamination of the glass surface by a single dirt feature, and their proportions are different. Therefore, the staff can set the coefficients corresponding to the area and gray value in advance and store them in the local storage medium in the drone. The drone calls the corresponding coefficients to perform weighted calculation to obtain the sub-contamination value of each dirt feature. The larger the sub-contamination value, the more serious the degree of contamination of the glass surface.
[0051] Sb determines the total contamination value of all contamination features based on the sub-contamination value of each contamination feature.
[0052] In this embodiment of the application, the UAV sums the sub-pollution values of all dirt features to obtain the total pollution value of all dirt features.
[0053] Sc is a value used to determine the degree of contamination on the glass surface based on the total contamination value, the first quantity, and the absolute value of the difference.
[0054] In this embodiment, the total contamination value, the first quantity of dirt characteristics, and the absolute value of the difference are all key factors affecting the degree of contamination on the glass surface. The total contamination value and the first quantity represent the influence of dirt adhering to the glass surface on the degree of contamination, while the absolute value of the difference represents the overall influence of dust contamination and dirt characteristics on the degree of contamination. Therefore, operators can set corresponding coefficients for the above three factors and store them in the local storage medium of the drone. The drone then uses these corresponding coefficients to perform a weighted calculation to obtain the degree of contamination on the glass surface. The degree of contamination determined by comprehensively considering the above three factors is more accurate.
[0055] One possible implementation of this application embodiment involves determining the spray flow rate and drone speed when spraying cleaning agent onto the glass surface based on the contamination level and dust deposition level values in step S104. Specifically, this includes steps S1041 (not shown in the figure) and S1042 (not shown in the figure). S1041, the score of the glass unit is determined based on the pollution level value, the dust deposition level value and their respective weights.
[0056] In this embodiment, both the contamination level and dust deposition level are key factors affecting the contamination level of the glass unit, influencing both the spray flow rate of the cleaning agent and the movement speed of the drone during cleaning agent application. The drone stores pre-set weights for each of these factors, and uses these weights to perform a weighted calculation of the contamination level and dust deposition level to obtain the score for the glass unit.
[0057] It should be noted that all coefficients and weights mentioned in the embodiments of this application can be adaptively modified and adjusted according to actual conditions and needs.
[0058] S1042, determine the target score interval from multiple preset score intervals. Each preset score interval corresponds to a preset spray flow rate and a preset moving speed. The preset spray flow rate and preset moving speed of the target score interval are defined as the spray flow rate and moving speed of the drone when spraying cleaning agent onto the glass surface. Assuming the pollution level is 16, the dust deposition level is 8, the weight corresponding to the pollution level is 0.6, and the weight corresponding to the dust deposition level is 0.4, then the score of the glass unit is 16 × 0.6 + 8 × 0.4 = 12.8.
[0059] In this embodiment, the drone compares the determined score with multiple preset score intervals to determine the target score interval. The preset spray flow rate and preset moving speed corresponding to each preset score interval are pre-set by relevant personnel based on extensive experimental testing or calculations. Using the preset spray flow rate and moving speed of the target score interval for cleaning agent spraying ensures more thorough and comprehensive application. Assuming there are three preset score intervals: interval a (0,5], interval b (5,10], and interval c (10,15]), the spray flow rate and moving speed corresponding to interval a are 0.03 m³ / s and 0.5 m / s, respectively; interval b is 0.05 m³ / s and 0.4 m / s, respectively; and interval c is 0.07 m³ / s and 0.3 m / s, respectively. If the drone determines that the score falls within interval c, then the drone determines the spray flow rate and moving speed of interval c as the spray flow rate and moving speed for spraying cleaning agent onto the current glass unit.
[0060] In one possible implementation of this application embodiment, the bottom image includes a first image of the horizontal plane at the bottom of the border and a second image of the vertical plane of the outer wall at the bottom of the border. Step S102, which determines the degree of dust deposition at the bottom of the border based on the bottom image, specifically includes steps S1026 (not shown in the figure), S1027 (not shown in the figure), S1028 (not shown in the figure), S1029 (not shown in the figure), and S10210 (not shown in the figure). S1026, Perform grayscale transformation on the first image to obtain the first grayscale image.
[0061] In this embodiment of the application, the bottom horizontal plane of the frame is the horizontal edge at the transition between the glass and the frame, where dust is prone to accumulate. The drone performs denoising processing on the first image, and then performs grayscale transformation on the denoised image to obtain the first grayscale image.
[0062] S1027, determine the first gray value histogram of the first grayscale image, and calculate the first similarity between the first gray value histogram and the first preset histogram.
[0063] The first preset histogram is the histogram of gray values when there is no dust deposition on the bottom horizontal plane of the border.
[0064] In this embodiment, the UAV generates a first grayscale histogram based on the grayscale value distribution in the first grayscale image, and then calculates a first similarity between the first grayscale histogram and a first preset histogram. The first preset histogram can be generated by manually cleaning the bottom horizontal surface of the frame of a glass unit, acquiring an image after cleaning and removing dust deposits, performing grayscale transformation to obtain a grayscale image, and then generating the first preset histogram based on the grayscale value distribution in the grayscale image. The first preset histogram represents the grayscale value histogram when there is no dust deposit on the bottom horizontal surface of the frame of all glass units. Specifically, the UAV calculates the Barcol distance or Euclidean distance between the first grayscale image and the first preset histogram, using the Barcol distance or Euclidean distance to represent the first similarity. A higher first similarity indicates less dust deposits along the horizontal edge.
[0065] S1028, perform grayscale transformation on the second image to obtain a second grayscale image.
[0066] S1029, determine the second grayscale value histogram of the second grayscale image, and calculate the second similarity between the second grayscale value histogram and the second preset histogram.
[0067] The second preset histogram is a histogram of gray values when there is no dust deposition on the vertical surface of the bottom outer wall of the frame.
[0068] In this embodiment of the application, the vertical surface of the bottom outer wall of the frame is a vertical outer surface perpendicular to the horizontal edge mentioned in step S1026. Dust may be adsorbed on this vertical surface under the action of static electricity, resulting in a certain degree of dust accumulation. During subsequent water rinsing and cleaning, the dust on this vertical surface will also mix with water, thus contaminating the cleaned glass below. The specific implementation of steps S1028 and S1029 can be referred to steps S1026 and S1027 and will not be repeated here.
[0069] S10210, determine the degree of dust deposition at the bottom of the border based on the first similarity and the second similarity.
[0070] In the embodiments of this application, both the first similarity and the second similarity are key factors characterizing dust deposition at the bottom of the frame. The UAV can sum the first similarity and the second similarity to obtain the degree of dust deposition at the bottom of the frame. It is more accurate to determine the overall degree of dust deposition at the bottom of the frame by comprehensively considering the dust deposition along the horizontal edges and vertical surfaces of the frame.
[0071] In one possible implementation of this application embodiment, step S106 (not shown in the figure) is included after step S105, wherein... S106, when the drone is detected to have reached the bottom of the glass unit's frame, the spray flow rate is increased.
[0072] In this embodiment of the application, when spraying cleaning agent onto the glass unit, the camera device on the drone captures an image of the glass unit. Based on this image, the drone determines whether it is about to reach the bottom of the frame. For example, when the bottom frame begins to appear in the image, it is determined that it is about to reach the bottom of the frame. At this time, the drone increases the spray flow rate based on the determined spray flow rate, thereby spraying more cleaning agent onto the bottom of the frame and allowing the cleaning agent to mix better with the dust remaining on the bottom of the frame. This makes the subsequent rinsing more thorough and prevents the dust-water mixture from contaminating the glass below. Specifically, the drone can store a preset value. The increased spray flow rate is obtained by adding the preset value to the determined spray flow rate.
[0073] One possible implementation of this application embodiment is that step S103, which determines the suction power of the dust collection component based on the degree of dust deposition, specifically includes step S1031 (not shown in the figure), wherein... S1031, retrieve the preset function used to calculate the suction value, and substitute the dust deposition level value into the preset function to obtain the suction value.
[0074] In this embodiment of the application, the preset function for determining the appropriate suction value based on the dust deposition level can be obtained in advance by relevant personnel through extensive testing and stored in the local storage medium of the drone. After the drone determines the dust deposition level, it can calculate the suction value by substituting the dust deposition level value into the preset function. For example, the preset function is y=150x+200, where y is the suction value in Pa, 150 is the proportionality coefficient, x is the dust deposition level, and 200 is the correction parameter. Taking the dust deposition level value 8 in step S1042 as an example, substituting 8 into the above preset function yields a suction value y of 1400 Pa.
[0075] The above embodiments describe a method for cleaning exterior walls based on drones from the perspective of process flow. The following embodiments describe a drone, and details are provided in the following embodiments.
[0076] This application provides an embodiment of a drone, such as... Figure 2 As shown, a drone may specifically include a drone body 1, a camera device 2 fixedly connected to the drone body 1, a dust collection component 3 disposed on the drone body 1, and a spraying component 4 fixedly connected to the drone body 1.
[0077] The camera device 2 can be one or two, and is set in front of the drone body 1 to capture images of the glass surface of the glass unit on the outer wall where the drone is located, as well as the bottom image of the glass unit's frame. The dust collection assembly 3 includes a motor 31 fixedly connected to the drone body 1, a gear 32 fixedly connected to the output shaft of the motor 31, a support frame 33 fixedly connected to the drone body 1, a rack 34 slidably connected to the support frame 33, a suction pipe 35 fixedly connected to the rack 34, and a blower 36 connected to the suction pipe 35.
[0078] Gear 32 meshes with rack 34. A long, narrow groove (not shown in the figure) is provided on the bottom surface of support frame 33. A retaining ring (not shown in the figure) can be fixedly connected to the bottom surface of rack 34, passing through the groove of support frame 33. The suction pipe 35 is fixed to support frame 33 via the retaining ring. When it is necessary to suck up dust from the bottom of the frame, motor 31 drives gear 32 to rotate. Gear 32 drives rack 34 to move laterally, which in turn drives suction pipe 35 to move laterally. This allows the drone to suck up dust from different locations on the bottom of the frame without moving laterally. Suction pipe 35 is connected to a ground blower 36 via a corrugated pipe or other flexible hose. The ground blower 36 operates to perform suction. The power of blower 36 can be adjusted to change the suction value. The spraying assembly 4 includes a spray nozzle 41 and a water pump 42. The power of the water pump 42 is also adjustable, thereby changing the spray flow rate. The nozzle of the spray nozzle 41 is fan-shaped, which increases the spraying area when spraying cleaning agent. The spray nozzle 41 is connected to the water pump 42 on the ground through a flexible hose such as a corrugated pipe. The water pump 42 is connected to a container holding the cleaning agent. The operation of the water pump 42 realizes the extraction and spraying of the cleaning agent. When rinsing off the cleaning agent later, the water pump 42 can be connected to a container holding water, thereby realizing the use of the spraying assembly 4 for rinsing.
[0079] Reference Figure 2 The suction pipe 35 has a notch 351 at its opening. When suctioning dust, part of the side wall of the suction pipe 35 at the notch 351 contacts the horizontal surface of the bottom frame, and another part contacts the vertical surface of the bottom frame, thereby achieving simultaneous suction of dust on the horizontal and vertical surfaces of the bottom frame and improving suction efficiency.
[0080] Furthermore, the suction pipe 35 and the blower 36 can also form a blowing assembly. When the blower 36 reverses, the suction pipe 35 blows out a high-speed airflow. After the drone identifies the dirt features on the glass surface, the drone can control the motor 31 to rotate and move itself to the dirt feature. At this time, the suction pipe 35 blows out a high-speed airflow, which causes the dirt to detach from the glass surface or become loose. This allows the dirt to better blend with the cleaning agent and be removed more easily.
[0081] This application provides an embodiment of a drone, such as... Figure 3 As shown, Figure 3 The drone shown includes a processor 51 and a memory 53. The processor 51 and memory 53 are connected, for example, via a bus 52. The camera device 2, blower 36, water pump 42, motor 31, and other components on the drone are also connected to the processor 51 via the bus 52. Optionally, the drone may also include a transceiver 54. It should be noted that in practical applications, the transceiver 54 is not limited to one type, and the structure of this drone does not constitute a limitation on the embodiments of this application.
[0082] Processor 51 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 51 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.
[0083] Bus 52 may include a pathway for transmitting information between the aforementioned components. Bus 52 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 52 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The symbol is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0084] The memory 53 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.
[0085] The memory 53 is used to store application code that executes the solution of this application, and its execution is controlled by the processor 51. The processor 51 is used to execute the application code stored in the memory 53 to implement the content shown in the foregoing method embodiments.
[0086] Figure 3 The drone shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of this application.
[0087] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the UAV described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0088] This application provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments. Compared with related technologies, this application acquires glass surface and bottom images, which record the specific contamination status of the glass surface and bottom edge. Therefore, the degree of contamination on the glass surface can be accurately determined based on the glass surface image. Similarly, the degree of dust deposition at the bottom of the edge can be accurately determined based on the bottom image. A higher dust deposition degree indicates more dust accumulation, which in turn indicates a greater likelihood that the dust-water mixture will flow along the edge to the lower glass surface during subsequent rinsing. Therefore, a stronger suction is needed to remove the dust from the bottom of the edge. The suction value determines the appropriate suction power of the vacuuming component, and the vacuuming component is controlled according to the determined suction power to better pick up the dust at the bottom of the frame. Based on the degree of contamination and dust deposition on the glass surface, the spray flow rate and the drone's movement speed are determined when spraying cleaning agent. This allows the cleaning agent to be sprayed more thoroughly and comprehensively onto the glass frame, making the stains dissolve and the dust mix more completely. During subsequent rinsing, the mixture of dust and water is less likely to flow to the lower glass surface, keeping the lower glass surface clean and reducing contamination of the cleaned glass below.
[0089] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.
[0090] The above description is only a partial embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.
Claims
1. A method for cleaning exterior walls based on unmanned aerial vehicles (UAVs), characterized in that, include: Acquire images of the glass surface of the glass unit on the outer wall at the location of the drone, as well as the bottom image of the glass unit's frame; The degree of contamination on the glass surface is determined based on the glass surface image, and the degree of dust deposition at the bottom of the frame is determined based on the bottom image. The suction power of the dust collection component is determined based on the dust deposition level value, and the operation of the dust collection component is controlled based on the suction power value to suck up the dust at the bottom of the frame. The spray flow rate and moving speed of the drone when spraying cleaning agent onto the glass surface are determined based on the pollution level value and the dust deposition level value. The drone is controlled to operate based on the stated movement speed and to spray cleaning agent according to the stated spray flow rate.
2. The method for cleaning exterior walls based on unmanned aerial vehicles according to claim 1, characterized in that, Determining the degree of contamination on the glass surface based on the glass surface image includes: Feature recognition is performed on the glass surface image to obtain the dirt features on the glass surface; Determine the area and gray value of each dirt feature, and count all dirt features to obtain the first quantity of all dirt. The glass surface image is subjected to grayscale transformation to obtain a surface grayscale image, and the average grayscale value of the surface grayscale image is determined. Determine the absolute value of the difference between the average gray value and the preset gray value; The degree of contamination on the glass surface is determined based on the area, gray value, first quantity, and absolute value of the difference of each dirt feature.
3. The method for cleaning exterior walls based on unmanned aerial vehicles according to claim 2, characterized in that, The determination of the degree of contamination on the glass surface based on the area, gray value, first quantity, and absolute value of the difference of each dirt feature includes: The sub-contamination value of each contamination feature is determined based on the area and gray value of each contamination feature; The total contamination value for all contamination features is determined based on the sub-contamination value of each contamination feature. The degree of contamination on the glass surface is determined based on the total contamination value, the first quantity, and the absolute value of the difference.
4. The method for cleaning exterior walls based on unmanned aerial vehicles according to claim 1, characterized in that, The determination of the spray flow rate and movement speed of the drone when spraying cleaning agent onto the glass surface based on the pollution level value and dust deposition level value includes: The score of the glass unit is determined based on the pollution level value, the dust deposition level value, and their respective weights. The target score interval is determined from multiple preset score intervals. Each preset score interval corresponds to a preset spray flow rate and a preset moving speed. The preset spray flow rate and preset moving speed of the target score interval are determined as the spray flow rate and moving speed of the drone when spraying cleaning agent onto the glass surface.
5. The method for cleaning exterior walls based on unmanned aerial vehicles according to claim 1, characterized in that, The bottom image includes a first image of the horizontal plane at the bottom of the border and a second image of the vertical plane of the outer wall at the bottom of the border. Determining the degree of dust deposition at the bottom of the border based on the bottom image includes: Perform a grayscale transformation on the first image to obtain a first grayscale image; Determine the first grayscale value histogram of the first grayscale image, and calculate the first similarity between the first grayscale value histogram and the first preset histogram, wherein the first preset histogram is the grayscale value histogram when there is no dust deposition on the bottom horizontal plane of the border; The second image is subjected to grayscale transformation to obtain a second grayscale image; Determine the second grayscale value histogram of the second grayscale image, and calculate the second similarity between the second grayscale value histogram and the second preset histogram, wherein the second preset histogram is the grayscale value histogram when there is no dust deposition on the vertical surface of the bottom outer wall of the frame; The degree of dust deposition at the bottom of the border is determined based on the first similarity and the second similarity.
6. The method for cleaning exterior walls based on unmanned aerial vehicles according to claim 4, characterized in that, The method further includes: When the drone is detected to have reached the bottom edge of the glass unit, the spray flow rate is increased.
7. The method for cleaning exterior walls based on unmanned aerial vehicles according to claim 1, characterized in that, Determining the suction power of the dust collection component based on the dust deposition level includes: A preset function for calculating the suction value is retrieved, and the dust deposition level value is substituted into the preset function to obtain the suction value.
8. A drone, characterized in that, It includes: The drone itself; A camera device is mounted on the drone body to capture images of the glass surface of the glass unit on the outer wall where the drone is located, as well as the bottom image of the glass unit's frame. A dust-collecting component, mounted on the drone body, is used to collect dust from the bottom of the frame. Spraying assembly for spraying cleaning agents; At least one processor, wherein the camera device, the dust collection assembly, and the spraying assembly are all communicatively connected to the at least one processor; Memory; At least one application, wherein the at least one application is stored in the memory and configured to be executed by the at least one processor, the at least one application being used to perform a drone-based exterior wall cleaning method according to any one of claims 1 to 7.
9. A drone according to claim 8, characterized in that, The dust collection assembly includes a motor mounted on the drone body, a gear mounted on the motor, a support frame mounted on the drone body, a rack mounted within the support frame, a dust collection pipe mounted on the rack, and a blower connected to the dust collection pipe.
10. A drone according to claim 9, characterized in that, The suction pipe has a notch at its port, which is located at the bottom of the suction pipe.