Vehicle-mounted large-particle-size rockfill water-saving cleaning system and method

Abstract

The invention relates to the technical field of rockfill cleaning, in particular to a vehicle-mounted large-particle-size rockfill water-saving cleaning system and method.The vehicle-mounted large-particle-size rockfill water-saving cleaning system comprises a spray head array unit, an image collecting unit, a gap recognizing unit and a central control unit, and the spray head array unit is configured to conduct penetrating type cleaning on target deep rockfill in a rockfill body; the image acquisition unit is configured to acquire an actual image of the rockfill; the gap identification unit is configured to position a gap area between rockfill and obtain three-dimensional position information of the mass center of the gap area; and the central control unit is configured to generate a cleaning task of the three-dimensional flushing working surface so as to drive the nozzle array unit to execute the cleaning task. Therefore, the problems that in the prior art, the nozzle angle is fixed, flexible adjustment cannot be achieved according to the loading capacity and shape of rockfill in the dumper, water flowing out of the nozzle impacts the surface of the rockfill, water flow splashes and escapes, the water resource utilization rate is low, cleaning dead corners exist, and the silt content of the rockfill is difficult to effectively control are solved.

Description

Technical Field

[0001] This application relates to the field of rockfill cleaning technology, and in particular to a vehicle-mounted water-saving cleaning system and method for large-diameter rockfill. Background Technology

[0002] In related technologies, the mud content of the rock used in rockfill concrete dams should not exceed 0.2%. As for the cleaning methods of the rockfill, there are fixed spray stations where the rockfill is unloaded from dump trucks and then spread and washed. This is not only time-consuming and energy-intensive, but also requires secondary transportation of the rockfill. During this process, the rockfill will collide and break with each other, producing small-diameter gravel, which affects the density of the rockfill concrete. Alternatively, by setting up a washing platform, water spray components and a water circulation treatment system, the rockfill in the dump truck can be washed directly, effectively avoiding the stone damage and pollution problems caused by secondary transportation.

[0003] However, in related technologies, the fixed nozzle angle cannot be flexibly adjusted according to the load and shape of the rock pile inside the dump truck. This results in the water from the nozzle directly impacting the surface of the rock pile, causing the water to splash and disperse, making it difficult to directly penetrate the gaps between the rocks. This leads to low water resource utilization and difficulty in effectively removing the clay adhering to the surface of the rock pile deep inside the rock pile, creating cleaning dead zones. Consequently, it is difficult to effectively control the mud content of the rock pile, which urgently needs to be addressed. Summary of the Invention

[0004] This application provides a vehicle-mounted water-saving cleaning system and method for large-diameter rockfill, which solves the problems in related technologies where the nozzle angle is fixed and cannot be flexibly adjusted according to the loading volume and shape of the rockfill in the dump truck. This results in the water from the nozzle directly impacting the surface of the rockfill, causing water to splash and disperse, making it difficult to directly penetrate the gaps between the rockfills, resulting in low water resource utilization. Furthermore, it is difficult to effectively remove the clay adhering to the surface of the rockfill deep within the rockfill, creating cleaning dead zones and making it difficult to effectively control the mud content of the rockfill.

[0005] The first aspect of this application provides a vehicle-mounted water-saving cleaning system for large-diameter rockfill, comprising: a nozzle array unit for penetrating cleaning of target deep rockfill in a vehicle compartment; an image acquisition unit for acquiring actual images of the rockfill from multiple perspectives; a gap recognition unit for locating gap regions between rockfills based on the actual images and obtaining three-dimensional position information of the centroid in the gap regions; and a central control unit for generating a cleaning task for a three-dimensional flushing operation surface based on the three-dimensional position information, thereby driving the nozzle array unit to execute the cleaning task.

[0006] Based on the above technical means, the embodiments of this application use a nozzle array unit for penetrating cleaning, which can directly act on the rocks inside the rockfill. In addition, the image acquisition unit acquires actual images from multiple perspectives, providing comprehensive and three-dimensional visual data support for gap identification. Furthermore, the gap identification unit accurately locates the gap area and obtains its three-dimensional position information, achieving accurate perception of the gaps in the rockfill and providing reliable location basis for cleaning. At the same time, the central control unit drives the nozzle array unit to perform the cleaning task of the three-dimensional flushing operation surface, abandoning flooding cleaning and achieving targeted and precise flushing. This effectively eliminates blind spots in rockfill cleaning, significantly improves the cleanliness of the deep rocks in the rockfill, meets strict mud content control standards, reduces ineffective water consumption from the source, greatly improves water resource utilization, reduces long-term operating costs, and achieves an upgrade from manual and extensive cleaning to intelligent and precise cleaning through full-process automated control, significantly improving operational efficiency.

[0007] Optionally, in one embodiment of this application, it further includes: a nozzle angle control unit, which generates a target angle command for the nozzle array unit according to the cleaning task, so as to drive multiple nozzles of the nozzle array unit to rotate to the corresponding target angle.

[0008] Based on the above technical means, the embodiments of this application can drive multiple nozzles to rotate to the corresponding target angle, thereby accurately adjusting the nozzle spray direction according to the cleaning task, allowing the water flow to be directly aimed at the gaps between the piles of rocks, washing the deep piles of rocks that are difficult to clean, greatly reducing water splashing and waste, improving the overall level of intelligence and resource utilization efficiency of the cleaning operation, and reducing the long-term operating cost of the system.

[0009] Optionally, in one embodiment of this application, the nozzle array unit further includes: a horizontal water pipe rack disposed at a preset height on the top surface of the carriage, wherein multiple nozzles are distributed on the water pipe rack; a vertical water pipe rack and a lifting drive mechanism, wherein the lifting drive mechanism adjusts the vertical height of the vertical water pipe rack relative to the carriage.

[0010] Based on the above technical means, the embodiments of this application form a three-dimensional flushing operation surface that surrounds the rock pile from the top and sides through a horizontal water pipe frame, a vertical water pipe frame and a lifting drive mechanism, so that the water flow can act more accurately on the deep parts of the rock pile, effectively eliminate cleaning blind spots, and improve the uniformity and cleanliness of flushing.

[0011] Optionally, in one embodiment of this application, the gap recognition unit is specifically used to preprocess and feature enhance the actual image to obtain a processed image that meets preset conditions; identify the pile area in the processed image to generate a pile binary map, and extract the potential gap area in the processed image to generate a gap binary map; perform multi-view image adaptation processing on the pile binary map and the gap binary map according to the acquisition type of the actual image, and perform morphological optimization processing on the gap binary map to determine the spatial relationship between the pile outline and the potential gap outline; and determine the gap area based on the spatial relationship to obtain the three-dimensional position information.

[0012] Based on the above technical means, this application embodiment constructs a mapping relationship between the image pixel coordinate system and the actual carriage coordinate system according to the perspective transformation matrix, thereby achieving a precise correspondence between image visual information and the physical space of the carriage. In addition, by converting the pixel coordinates of the centroid into two-dimensional and three-dimensional spatial position information of the carriage, the actual spatial orientation of the gap area between the piled stones inside the carriage is accurately located, improving the accuracy of the gap identification and the reliability of spatial positioning, enhancing the pertinence of subsequent operations, and reducing water waste.

[0013] Optionally, in one embodiment of this application, the gap recognition unit includes: a generation subunit, used to obtain a perspective transformation matrix based on the corner feature points of the top opening of the carriage; an establishment subunit, used to establish a mapping relationship between the image pixel coordinate system and the actual carriage coordinate system based on the top surface of the boulders body, based on the perspective transformation matrix; and a calculation subunit, used to calculate the pixel coordinates of the centroid based on the mapping relationship and the gap region, and convert the pixel coordinates into two-dimensional coordinates of the carriage, so as to generate the three-dimensional position information by combining the distance coordinate value determined based on the average distance of the top or side surface of the boulders body.

[0014] Based on the above technical means, this application embodiment improves image clarity and regional differentiation by preprocessing and feature enhancement of actual images, providing a foundation for subsequent recognition. Then, by identifying and extracting the rockfill area and potential gap area, the initial screening of gap area is completed. Subsequently, according to the acquisition type of the actual image, the binary image of the rockfill and the binary image of the gap are subjected to multi-view image adaptation processing to improve scene applicability. Furthermore, the binary image of the gap is subjected to morphological optimization processing to ensure the accuracy of contour recognition, thereby effectively determining the gap area and obtaining its three-dimensional position information. This significantly enhances the targeting of subsequent operations, reduces water waste, and improves the overall operation efficiency and intelligence level.

[0015] Optionally, in one embodiment of this application, the conversion formula for the two-dimensional coordinates of the carriage is: , , in, For the horizontal component in homogeneous coordinates. The vertical component in the homogeneous coordinates. The scaling factor in the homogeneous coordinates. This is the perspective transformation matrix. The horizontal coordinate in the pixel coordinates of the centroid. The vertical coordinate in the pixel coordinates of the centroid. The horizontal coordinate in the two-dimensional coordinate system of the carriage is [the horizontal coordinate]. The vertical coordinate in the two-dimensional coordinate system of the carriage is denoted as .

[0016] Based on the above technical means, the embodiments of this application can accurately convert the pixel positions visually recognized in the top or side images of the rock pile into two-dimensional spatial coordinates in the actual car body coordinate system, and form three-dimensional position information from the average distance coordinates measured by the ranging sensor. This effectively eliminates the positional deviation between the image and the actual space, improves the positioning accuracy and scene adaptability of subsequent operations, and helps subsequent operation links to more efficiently match the actual stacking state of the rock pile.

[0017] A second aspect of this application provides a water-saving cleaning method for large-diameter rockfill piles mounted on a vehicle, comprising: acquiring actual images of the rockfill piles inside the vehicle from multiple perspectives; locating gap regions between the rockfill piles based on the actual images from multiple perspectives, and obtaining three-dimensional position information of the gap regions; generating a cleaning task for a three-dimensional flushing operation surface based on the three-dimensional position information of the gap regions, so as to execute the cleaning task.

[0018] Based on the above technical means, the embodiments of this application can directly act on the internal rockfill through penetrating cleaning. In addition, by acquiring actual images from multiple perspectives and information obtained by ranging sensors, comprehensive and three-dimensional visual data support is provided for the identification of gaps in the rockfill. Furthermore, by accurately locating the gap area and obtaining its three-dimensional position information, accurate perception of the gaps in the rockfill is achieved, providing a reliable location basis for cleaning. At the same time, by performing the cleaning task of three-dimensional flushing operation, the flooding cleaning method is abandoned, and targeted and precise flushing is achieved. This not only effectively eliminates blind spots in rockfill cleaning and significantly improves the cleanliness of the deep rockfill, meeting strict mud content control standards, but also reduces ineffective water consumption from the source, greatly improves water resource utilization, and reduces long-term operating costs. Through full-process automated control, the cleaning operation is upgraded from manual and extensive to intelligent and precise, significantly improving operational efficiency.

[0019] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the vehicle-mounted large-diameter rockfill water-saving cleaning method as described in the above embodiments.

[0020] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described vehicle-mounted large-diameter rockfill water-saving cleaning method.

[0021] A fifth aspect of this application provides a computer program product, including a computer program that, when executed, implements the above-described vehicle-mounted large-diameter rockfill water-saving cleaning method.

[0022] This application embodiment utilizes penetrating cleaning to reach the internal rockfill. Furthermore, by acquiring actual images from multiple perspectives, it provides comprehensive and three-dimensional visual data support for identifying gaps in the rockfill. Additionally, by accurately locating gap areas and acquiring their three-dimensional position information, it achieves precise perception of the rockfill gaps, providing a reliable location basis for cleaning. Simultaneously, by performing a three-dimensional flushing operation, it abandons flood-style cleaning and achieves targeted and precise flushing. This effectively eliminates blind spots in rockfill cleaning, significantly improves the cleanliness of deep rockfill layers, meets stringent mud content control standards, reduces ineffective water consumption at the source, greatly improves water resource utilization, lowers long-term operating costs, and upgrades cleaning operations from manual and extensive to intelligent and precise through fully automated control, significantly improving operational efficiency. This solves the problem in related technologies where the nozzle angle is fixed and cannot be flexibly adjusted according to the load and shape of the rock pile in the dump truck. This results in the water from the nozzle directly impacting the surface of the rock pile, causing water to splash and disperse, making it difficult to directly penetrate the gaps between the rocks, leading to low water resource utilization. Furthermore, it is difficult to effectively remove the clay attached to the deep layers of the rock pile, resulting in cleaning dead zones and difficulty in effectively controlling the mud content of the rock pile.

[0023] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0024] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a block diagram of a vehicle-mounted large-diameter rockfill water-saving cleaning system provided according to an embodiment of this application; Figure 2 This is a top view of a wire mesh cage being loaded into a dump truck according to an embodiment of this application; Figure 3This is a perspective view of a horizontal and vertical water pipe rack provided according to an embodiment of this application; Figure 4 This is a schematic diagram of the hardware arrangement of an image acquisition unit according to an embodiment of this application; Figure 5 This is a schematic diagram of the algorithm flow for configuring a gap recognition unit according to an embodiment of this application; Figure 6 This is a schematic diagram of interstitial targeted flushing according to an embodiment of this application; Figure 7 This is a schematic diagram of the principle of a vehicle-mounted large-diameter rockfill water-saving cleaning system according to an embodiment of this application; Figure 8 This is a flowchart of a vehicle-mounted large-diameter rockfill water-saving cleaning method according to an embodiment of this application; Figure 9 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application.

[0025] Figure label: 10-Vehicle-mounted large-diameter rockfill water-saving cleaning system; 100-Sprayer array unit, 200-Image acquisition unit, 300-Gap recognition unit, 400-Central control unit; 201-Dump truck, 202-Net cage box; 301-Adjustable angle sprayer, 302-Horizontal and vertical water pipe rack; 401-Camera, 402-Distance sensor, 403-Rockfill, 404-Clay, 405-Truck floor, 406-Rockfill body; 901-Memory, 902-Processor, 903-Communication interface. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0027] The following description, with reference to the accompanying drawings, illustrates a vehicle-mounted water-saving cleaning system and method for large-diameter rockfill piles according to embodiments of this application. Addressing the issues raised in the background art where the nozzle angle is fixed and cannot be flexibly adjusted according to the rockfill load and shape within the dump truck, resulting in water directly impacting the rockfill surface and splashing away, making it difficult to penetrate the gaps between the rocks, leading to low water utilization and difficulty in effectively removing clay from deep within the rockfill, creating cleaning dead zones and hindering effective control of the mud content, this application provides a vehicle-mounted water-saving cleaning system for large-diameter rockfill piles. In this system, penetrating cleaning can directly target deep-seated rocks within the pile. Furthermore, by acquiring actual images from multiple perspectives, a comprehensive and three-dimensional visual system is provided for gap identification. Supported by sensory data, and by precisely locating gap areas and acquiring their three-dimensional position information, accurate perception of the gaps in the rockfill is achieved, providing a reliable location basis for cleaning. Simultaneously, by executing a three-dimensional flushing operation, the system abandons flooding cleaning and achieves targeted and precise flushing. This effectively eliminates blind spots in rockfill cleaning, significantly improves the cleanliness of deep rockfill layers, meets strict mud content control standards, reduces ineffective water consumption at the source, greatly improves water resource utilization, and lowers long-term operating costs. Furthermore, through fully automated control, the cleaning operation is upgraded from manual and extensive to intelligent and precise, significantly improving operational efficiency. This solves the problems of fixed nozzle angles in related technologies, which cannot be flexibly adjusted according to the load and shape of the rockfill in the dump truck. This results in water directly impacting the rockfill surface, splashing and scattering, making it difficult to directly penetrate the gaps in the rockfill, leading to low water resource utilization, difficulty in effectively removing clay adhering to deep rockfill layers, the existence of cleaning dead zones, and difficulty in effectively controlling the mud content of the rockfill.

[0028] Specifically, Figure 1 This is a block diagram of a vehicle-mounted large-diameter rockfill water-saving cleaning system provided according to an embodiment of this application.

[0029] like Figure 1 As shown, the vehicle-mounted large-diameter rockfill water-saving cleaning system 10 may include, but is not limited to: a nozzle array unit 100, an image acquisition unit 200, a gap recognition unit 300, and a central control unit 400.

[0030] The nozzle array unit 100 is used to perform penetrating cleaning of the target deep rock pile in the rock pile of the carriage.

[0031] In the embodiments of this application, the riprap body can be understood as a collection of stones loaded in a carriage, which is formed by the natural or regular stacking of multiple large-diameter riprap stones. It contains a large number of gap areas between individual riprap stones and can be divided into surface riprap and deep riprap according to the stacking position.

[0032] In addition, the target deep rockfill can be understood as individual rockfills located below the surface rockfill and partially or completely obscured by other rockfills. They are usually located in the middle and lower part of the rockfill and are difficult to clean effectively due to the obstruction of the surface rockfill.

[0033] It is understandable that the nozzle array unit 100 constitutes the execution layer of the system. The nozzle array unit 100 will be described in detail below.

[0034] Specifically, in one embodiment of this application, the nozzle array unit 100 includes: a horizontal water pipe frame disposed at a preset height on the top surface of the carriage, wherein multiple nozzles are distributed on the water pipe frame; a vertical water pipe frame and a lifting drive mechanism, wherein the lifting drive mechanism adjusts the vertical height of the vertical water pipe frame relative to the carriage.

[0035] In actual operation, the nozzle array unit 100 may include, but is not limited to, a horizontal water pipe rack installed on the top surface of the carriage, and a vertical water pipe rack controlled by a lifting drive mechanism (such as a hydraulic or electric cylinder) that can descend vertically and precisely embed itself into the side gaps of the riprap cage. Adjustable-angle nozzles are distributed and installed on the water pipe racks. The synergistic effect of the horizontal and vertical water pipe racks forms a three-dimensional washing surface surrounding the riprap from the top and sides (the horizontal water pipe rack is typically 50 cm to 80 cm above the top surface of the riprap in the carriage), ensuring penetrating cleaning of the target deep layers of riprap within the riprap body.

[0036] For example, such as Figure 2 The diagram shows a top view of the wire mesh cages being installed in a dump truck, where 201 represents the dump truck and 202 represents the wire mesh cage. In practical applications, the two wire mesh cages are typically installed side-by-side inside the dump truck's cargo compartment. To achieve three-dimensional targeted washing, a gap of 15–20 cm is maintained between the wire mesh cages, and a gap of 10–15 cm is maintained between the wire mesh cages and the side panels of the cargo compartment. These gaps provide the necessary physical space for the subsequent embedding of the nozzle array.

[0037] like Figure 3 The diagram shows a three-dimensional representation of the horizontal and vertical water pipe supports. 301 represents an adjustable-angle nozzle, and 302 represents the horizontal and vertical water pipe supports themselves. The horizontal and vertical water pipe supports are rectangular frames that provide stable support for the nozzles. 36 nozzles are mounted on the horizontal water pipe support, and 100 nozzles are mounted on the vertical water pipe support, together forming a three-dimensional rinsing surface.

[0038] This embodiment of the application forms a three-dimensional flushing operation surface that surrounds the rock pile from the top and sides by using a horizontal water pipe frame, a vertical water pipe frame, and a lifting drive mechanism. This allows the water flow to act more precisely on the deep parts of the rock pile, effectively eliminating cleaning blind spots and improving the uniformity and cleanliness of the flushing.

[0039] Image acquisition unit 200 is used to acquire actual images of the rockfill body from multiple perspectives.

[0040] In actual operation, the image acquisition unit 200 may include, but is not limited to, multiple waterproof industrial cameras. These waterproof industrial cameras are distributed and installed on the water pipe rack of the nozzle array unit 100 to acquire actual images of the piled stones inside the carriage from above and the side, and transmit the actual images to the gap recognition unit 300 for processing.

[0041] In addition, the image acquisition unit 200 also includes multiple ranging sensors, which are distributed on the water pipe frame of the nozzle array unit 100 to measure the average distance from the ranging sensor to the top or side of the rock pile and transmit the data to the gap recognition unit 300 for coordinate transformation.

[0042] In addition, waterproof industrial cameras and rangefinders can be mounted at the same height or on the same plane on the water pipe rack.

[0043] For example, such as Figure 4 The diagram shows the hardware layout of the image acquisition unit. 401 is the camera, 402 is the range sensor, 301 is the adjustable-angle nozzle, 403 is the riprap, 404 is the clay, 405 is the truck bed, and 406 is the riprap body. The camera and range sensor can be fixed at the midpoint between the horizontal and vertical water pipe supports to ensure full coverage of the riprap surface.

[0044] The gap recognition unit 300 is used to locate the gap area between the rocks based on the actual image and obtain the three-dimensional position information of the centroid in the gap area.

[0045] In the embodiments of this application, the gap area can be understood as the blank space area without the solidity of the piled stones formed between individual piled stones and between the piled stones and the inner wall of the carriage during the process of piling stones in the carriage. Clay and other impurities are easily attached to the gap area.

[0046] In addition, three-dimensional position information can be understood as the three-dimensional coordinate data of the target (such as the centroid of the gap between the boulders) in the actual physical space of the carriage. For example, the two-dimensional coordinates of the carriage obtained by perspective transformation, combined with the average distances from multiple sensors to the top or side of the boulders measured by the distance measuring sensor, constitute the three-dimensional spatial coordinates.

[0047] In actual execution, the gap recognition unit 300 can recognize the actual images transmitted by the image acquisition unit 200 from above and to the side of the rock pile to locate the gap area between the rocks and output its three-dimensional position information (such as three-dimensional spatial coordinates).

[0048] Specifically, in one embodiment of this application, the gap recognition unit 300 is specifically used to preprocess and enhance the features of the actual image to obtain a processed image that meets preset conditions; identify the pile area in the processed image to generate a pile binary map, and extract the potential gap area in the processed image to generate a gap binary map; perform multi-view image adaptation processing on the pile binary map and the gap binary map according to the acquisition type of the actual image, and perform morphological optimization processing on the gap binary map to determine the spatial relationship between the analyzed pile contour and the potential gap contour; and determine the gap area based on the spatial relationship to obtain three-dimensional position information.

[0049] In actual implementation, such as Figure 5 As shown, the gap recognition unit 300 can be configured to perform the following processing flow.

[0050] Step S501: Image preprocessing and feature enhancement.

[0051] In this embodiment, the actual image collected is converted from the BGR (Blue Green Red) color space to the HSV (Hue Saturation Value) color space to enhance the color contrast between the pile and the gap area.

[0052] Step S502: Parallel segmentation of dual regions.

[0053] In this embodiment, the image is segmented in parallel based on a certain HSV threshold range. Specifically, based on the optical reflection characteristics of the boulders, a first HSV threshold range (e.g., H: 0-180, S: 0-80, V: 30-220) is used to identify the boulders region and generate a boulders binary map. In addition, based on the shadow characteristics of the gaps, a second HSV threshold range (e.g., H: 0-180, S: 0-255, V: 0-60) is used to extract potential gap regions and generate a gap binary map.

[0054] Step S503: Multi-view image adaptation processing.

[0055] In this application embodiment, adaptation processing is performed for the different characteristics of top and side images. For side images, due to their non-vertical top-down shooting angle, a correction matrix pre-calibrated for this viewpoint is first used to eliminate perspective distortion, making it closer to the geometric relationship of the top view. At the same time, considering that the side is usually less lit and has more significant shadow areas, the upper limit of the V (luminance) component in the second HSV threshold range (used for gap extraction) can be dynamically adjusted to 80 to optimize the segmentation effect.

[0056] Step S504: Morphological optimization.

[0057] In this embodiment, morphological opening and closing operations are sequentially performed on the binary gap map to improve quality. First, an opening operation is performed using a structuring element of a first size (e.g., 8×8 pixels to 12×12 pixels) to eliminate noise interference. Then, a closing operation is performed using a larger structuring element of a second size (e.g., 18×18 pixels to 22×22 pixels) to connect adjacent gap regions.

[0058] Step S505: Contour relationship verification and gap confirmation.

[0059] In this embodiment, the extracted riprap contours and potential gap contours are analyzed for spatial relationship, and a hierarchical algorithm is used to eliminate false gap regions. The algorithm includes: calculating the centroid coordinates of each potential gap contour; when the centroid is located inside any single riprap contour, it is determined to be a false gap; calculating the area overlap rate between each potential gap contour and all riprap contours; when the overlap rate with any single riprap contour exceeds a certain threshold (e.g., 60%), it is determined to be a false gap. This step effectively distinguishes between surface stains and textures on the riprap and the actual effective gaps between the riprap.

[0060] Step S506: Contour feature analysis and screening.

[0061] In this embodiment, a certain area threshold (e.g., 80 cm²) is used. 2 Effective filtration is performed to remove smaller gaps that are difficult for water to penetrate.

[0062] Step S507: Coordinate transformation and 3D reconstruction.

[0063] In this embodiment, the centroid coordinates of each effective contour are calculated, and then converted to three-dimensional spatial coordinates in the actual carriage coordinate system using a coordinate mapping algorithm before being output.

[0064] Furthermore, in one embodiment of this application, the gap recognition unit 300 includes: a generation subunit, used to obtain a perspective transformation matrix based on the corner feature points of the top opening of the carriage; an establishment subunit, used to establish a mapping relationship between the image pixel coordinate system and the actual carriage coordinate system based on the top surface of the boulders body, based on the perspective transformation matrix; and a calculation subunit, used to calculate the pixel coordinates of the centroid based on the mapping relationship and the gap area, and convert the pixel coordinates into two-dimensional coordinates of the carriage, so as to generate three-dimensional position information by combining the distance coordinate value determined based on the average distance of the top or side surface of the boulders body.

[0065] In the embodiments of this application, the corner feature points can be understood as the four physical vertices at the edge of the opening on the roof of the carriage. By obtaining the correspondence between their image pixel coordinates and the actual carriage coordinates, the perspective transformation matrix can be constructed in the embodiments of this application.

[0066] Furthermore, the perspective transformation matrix can be understood as a matrix obtained by calibrating the correspondence between the image pixel coordinates of the corner feature points of the carriage and the actual carriage coordinates, and is used to establish the spatial mapping rules between the image pixel coordinate system and the actual carriage coordinate system.

[0067] In addition, the image pixel coordinate system can be understood as a coordinate system established based on the actual image. For example, it uses the image pixels as the unit of measurement, takes a certain vertex of the image (such as the upper left corner) as the origin, and its horizontal direction corresponds to the pixel column of the image, and its vertical direction corresponds to the pixel row of the image. It is used to describe the two-dimensional position of the target (such as the centroid) in the image within the image frame.

[0068] In addition, the actual car body coordinate system can be understood as a coordinate system established based on the actual physical space of the car body. For example, the origin and coordinate axis directions are determined by taking the open top surface of the car body as the reference plane, and the unit of measurement is physical length (such as centimeters or meters). It is used to describe the two-dimensional position of the target (such as the centroid of the gap between the rubble) in the actual physical space.

[0069] In actual implementation, the embodiments of this application can use the 3×3 perspective transformation matrix (homography matrix) H obtained in advance by calibrating the feature points at the four corners of the carriage to establish the mapping relationship between the image pixel coordinate system and the actual carriage coordinate system based on the opening of the carriage top surface.

[0070] Furthermore, in this embodiment of the application, for the identified effective gap, the pixel coordinates p(u,v) of its centroid are calculated, and the two-dimensional coordinates P(X,Y) of the carriage are obtained through a conversion formula.

[0071] Optionally, in one embodiment of this application, the transformation formula for the two-dimensional coordinates of the carriage may be, but is not limited to, the following: , , in, For the horizontal component in homogeneous coordinates. This represents the vertical component in homogeneous coordinates. The scaling factor in homogeneous coordinates. This is the perspective transformation matrix. The horizontal coordinate in the pixel coordinates of the centroid. The vertical coordinate of the centroid in pixel coordinates. The horizontal coordinate in the two-dimensional coordinate system of the carriage. The vertical coordinate is the two-dimensional coordinate of the carriage.

[0072] It is understood that, according to the embodiments of this application, the homogeneous coordinates corresponding to the pixel coordinates of the centroid of the effective gap can be obtained by using the perspective transformation matrix and matrix multiplication based on the pixel coordinates of the centroid of the effective gap. Then, the horizontal and vertical components of the homogeneous coordinates are divided by the scaling factor to obtain the two-dimensional coordinates of the carriage.

[0073] The embodiments of this application can accurately convert the pixel positions obtained by visual recognition in the image into two-dimensional spatial coordinates in the actual carriage coordinate system, effectively eliminating the positional deviation between the image and the actual space, improving the positioning accuracy and scene adaptability of subsequent operations, and helping subsequent operation stages to more efficiently match the actual stacking state of the rock pile.

[0074] Furthermore, in this embodiment, a distance measuring sensor (such as an ultrasonic or laser sensor) installed on a water pipe rack is used to measure the average distance from the distance measuring sensor to the top or side of the rock pile in real time. This average distance is used as the unified Z-coordinate for all identification gaps to form a dynamic distance adaptive mechanism for the system. This allows the pitch angle (θ) and yaw angle (φ) calculated subsequently to dynamically adapt to the actual accumulation of the rock pile each time it is loaded, significantly improving the accuracy of nozzle angle alignment. This is one of the intelligent features for achieving efficient penetration cleaning.

[0075] Furthermore, in this embodiment, the two-dimensional plane coordinates (X, Y) obtained by perspective transformation are combined with the Z coordinates obtained by the ranging sensor to form complete three-dimensional spatial coordinates (X, Y, Z) in a unified carriage coordinate system, thereby outputting structured gap information including the three-dimensional coordinates (X, Y, Z) of the gap center to the central control unit 400.

[0076] Based on the description of other embodiments, exemplarily, in this application embodiment, four ultrasonic ranging sensors are symmetrically installed on a horizontal water pipe frame, and 16 ultrasonic ranging sensors are installed on a vertical water pipe frame to form a comprehensive monitoring network for the surface of the rock pile. The ultrasonic ranging sensors measure the distance from the surface of the rock pile in real time at a frequency of 10 Hz, and convert it into a height value based on the surface formed by the ranging sensors installed on the same side of the horizontal or vertical water pipe frame through geometric relationships. Furthermore, the central control unit 400 performs weighted averaging processing on the data from multiple sensors to eliminate the influence of local undulations and obtain the average distance Z_avg between the ranging sensors and the top or side surface of the rock pile. Z_avg is used as the unified Z coordinate of all identification gaps, so that the angle calculation can dynamically adapt to the actual accumulation situation of each loading of rock pile.

[0077] This application embodiment constructs a mapping relationship between the image pixel coordinate system and the actual carriage coordinate system based on the perspective transformation matrix, thereby achieving a precise correspondence between image visual information and the physical space of the carriage. In addition, by converting the pixel coordinates of the centroid into two-dimensional and three-dimensional position information of the carriage, the actual spatial orientation of the gap area inside the carriage is accurately located, improving the accuracy of gap recognition and the reliability of spatial positioning, enhancing the pertinence of subsequent operations, and reducing water waste.

[0078] In summary, this application embodiment improves image clarity and regional differentiation by preprocessing and feature enhancement of actual images, providing a foundation for subsequent recognition. Then, by identifying and extracting the rockfill area and potential gap area, the initial screening of gap areas is completed. Subsequently, based on the acquisition type of the actual image, the binary images of the rockfill and gap are subjected to multi-view image adaptation processing to improve scene applicability. Furthermore, the morphological optimization processing of the gap binary image ensures the accuracy of contour recognition, thereby effectively determining the gap area and obtaining its three-dimensional position information. This significantly enhances the targeting of subsequent operations, reduces water waste, and improves overall operational efficiency and intelligence.

[0079] The central control unit 400 is used to generate a cleaning task for a three-dimensional flushing operation surface based on the three-dimensional position information of the gap area, so as to drive the nozzle array unit 100 to perform the cleaning task.

[0080] It is understandable that the central control unit 400, as the command center of the system, is in communication connection with the image acquisition unit 200, the gap recognition unit 300, the nozzle angle control unit, and the lifting drive mechanism of the nozzle array unit 100.

[0081] In actual execution, the central control unit 400 is responsible for the process control of the entire washing operation, which may include, but is not limited to: after the vehicle enters the washing station and completes system initialization, the central control unit 400 sends a descent command to the lifting drive mechanism of the nozzle array unit 100 to control the vertical water pipe rack to descend precisely to a certain working height and embed into the side gap of the rockfill cage box to form a complete three-dimensional washing operation surface; the central control unit 400 receives the three-dimensional position information of all gaps output by the gap identification unit 300 and executes washing decisions and task allocation.

[0082] Specifically, the central control unit 400 can generate control commands based on the following intelligent allocation strategy: In this embodiment of the application, a car body coordinate system (O-XYZ) is established. The car body coordinate system can be defined as follows: the origin O is the geometric center of the open top surface of the car body, and the horizontal plane in which it is located is the O-XY plane; the Z-axis is perpendicular to the O-XY plane and points downwards, and the zero point of the Z-axis is determined by the mounting plane formed by the four ranging sensors fixed on the horizontal water pipe frame.

[0083] In this embodiment, for each nozzle, within its effective working radius of 30cm to 50cm, a nearest neighbor search algorithm is used to dynamically lock the nearest unallocated gap center point as its exclusive target point. This strategy simplifies the system control logic and improves decision-making speed and system reliability while ensuring flushing effect.

[0084] Once an effective gap center point is assigned to a nozzle, this embodiment immediately removes it from the set of assignable target points to ensure that each gap is targeted by only one nozzle in a single flushing cycle, thus avoiding water waste and interference between flushing streams.

[0085] For each assigned target point, the central control unit 400 calculates the target pitch angle (θ) and target yaw angle (φ) required by the corresponding nozzle based on the spatial geometry model.

[0086] In this embodiment of the application, the pitch angle (θ) is defined as the angle between the nozzle axis and the horizontal plane (O-XY plane), and its expression may be, but is not limited to, as follows: , in, and These represent the coordinate deviations between the centroid of the target rockfill gap and the nozzle in the horizontal plane (O-XY plane). This is the vertical height difference between the ranging sensor and the target point's Z coordinate (i.e., the average distance between the ranging sensor and the top surface of the rockfill).

[0087] Yaw angle (φ) is defined as the angle between the projection of the nozzle axis onto the horizontal plane and the X-axis of the carriage coordinate system. Its expression can be, but is not limited to, as follows: , in, The gap direction angle; This is a direction correction factor (with a value range of, for example, 0.2-0.5) calibrated through hydrodynamic experiments, used to compensate for errors in the parabolic curve and refraction of the water flow; and These represent the coordinate deviations between the centroid of the target rockfill gap and the nozzle in the horizontal plane (O-XY plane).

[0088] After the washing operation is completed, the central control unit 400 sends an upward command to the lifting drive mechanism to control the vertical water pipe rack to rise smoothly to the initial safe position, preparing for the vehicle to leave.

[0089] Optionally, in one embodiment of this application, it further includes: a nozzle angle control unit, which generates a target angle command for the nozzle array unit according to the cleaning task, so as to drive multiple nozzles of the nozzle array unit to rotate to the corresponding target angle.

[0090] In the embodiments of this application, the target angle can be understood as the angle parameter of the nozzle rotation generated by the nozzle angle control unit according to the cleaning task, which may include, but is not limited to, pitch angle and yaw angle. The pitch angle can be understood as the angle of the nozzle rotation around the horizontal axis, used to adjust the nozzle to align with the gap between the pile of rocks upward and downward. The yaw angle can be understood as the angle of the nozzle rotation around the vertical axis, used to adjust the nozzle to align with the gap between the pile of rocks to the left and right.

[0091] In addition, the target angle command can be understood as a command generated by the nozzle angle control unit based on the cleaning task, used to control the nozzle array unit to adjust the nozzle spray direction.

[0092] In actual operation, the nozzle angle control unit receives the target angle command from the central control unit 400 and, through its mechanical structure and control system, converts the command into the precise spatial orientation of the nozzle.

[0093] The mechanical structure employs a dual-axis series layout: a yaw axis system fixed to the base drives the nozzle to rotate horizontally around the vertical axis via a first stepper motor and transmission mechanism (such as a worm gear mechanism), achieving full-range adjustment of the yaw angle (φ); a pitch axis system mounted on the moving parts of the yaw axis system drives the nozzle to pitch around the horizontal axis via a second stepper motor and transmission mechanism, achieving independent adjustment of the pitch angle (θ). Furthermore, each axis system is equipped with sealing components to prevent water mist intrusion and ensure reliability in humid environments.

[0094] In addition, the control system receives the target angle (θ, φ) command issued by the central control unit 400, and based on the real-time feedback of the angle sensor integrated on the shaft system, drives the stepper motor to move through the closed-loop servo control algorithm until the actual direction of the nozzle is consistent with the target command, thereby achieving precise alignment of the target gap.

[0095] For example, such as Figure 6 The diagram illustrates a targeted flushing system for gaps, where 301 is an adjustable-angle nozzle, 403 is rockfill, and 404 is clay. The control system employs an embedded microcontroller that receives the target angle (θ, φ) (where θ is 0° to ±45° and φ is 0° to ±45°) from the central control unit 400. It reads real-time feedback signals from the encoder and tilt sensor integrated on the shaft system and drives the stepper motor through a closed-loop servo control algorithm until the nozzle's actual pointing direction matches the target command, achieving a positioning accuracy of ±0.5°, ensuring the nozzle's spray axis is precisely aligned with the center of the target gap.

[0096] This application embodiment drives multiple nozzles to rotate to the corresponding target angle, which can accurately adjust the nozzle spray direction according to the cleaning task, so that the water flow is directly aimed at the deep parts of the rock pile and other difficult-to-clean areas, greatly reducing water splashing and waste, improving the overall intelligence level and resource utilization efficiency of the cleaning operation, and reducing the long-term operating cost of the system.

[0097] The principle of the vehicle-mounted large-diameter rockfill water-saving cleaning system proposed in this application is illustrated below with reference to a specific embodiment.

[0098] like Figure 7 As shown, workers secure the wire mesh cages loaded with boulders inside the dump truck's cargo compartment to ensure a lateral working clearance, and then the dump truck drives into the washing station and initializes the system.

[0099] Furthermore, the central control unit controls the lifting drive mechanism of the nozzle array unit, causing the vertical water pipe rack to descend and embed into the side working gap, forming a three-dimensional washing working surface. In addition, the image acquisition unit acquires actual images of the rock pile surface through distributed industrial cameras, and the gap recognition unit obtains the three-dimensional position information of the rock pile gap through coordinate transformation and distance data obtained by the distance sensor.

[0100] Furthermore, the central control unit receives the three-dimensional position information of the gap from the gap identification unit, calculates the target pitch angle θ and yaw angle φ of each nozzle based on the intelligent allocation strategy, and generates a target angle command.

[0101] Furthermore, the nozzle angle control unit receives the target angle command from the central control unit, and drives the nozzle to precisely align with the target gap through a dual-axis serial mechanical structure and a closed-loop servo control system, performing three-dimensional penetrating targeted flushing.

[0102] Furthermore, the central control unit controls the lifting drive mechanism of the nozzle array unit, so that after the vertical water pipe frame returns to the initial safe position, the dump truck drives away from the work station.

[0103] The vehicle-mounted large-diameter rockfill water-saving cleaning system proposed in this application can penetrate deep into the rockfill body through penetrating cleaning. In addition, by acquiring actual images from multiple perspectives, it provides comprehensive and three-dimensional visual data support for gap identification. Furthermore, by accurately locating gap areas and acquiring their three-dimensional position information, it achieves accurate perception of rockfill gaps, providing reliable location basis for cleaning. At the same time, by performing the cleaning task of three-dimensional flushing operation, it abandons flooding cleaning and achieves targeted and precise flushing, which not only effectively eliminates blind spots in rockfill cleaning and significantly improves the cleanliness of deep rockfill, meeting strict mud content control standards, but also reduces ineffective water consumption from the source, greatly improves water resource utilization, and reduces long-term operating costs. Through full-process automated control, it realizes the upgrade of cleaning operations from manual and extensive to intelligent and precise, significantly improving operation efficiency. This solves the problem in related technologies where the nozzle angle is fixed and cannot be flexibly adjusted according to the load and shape of the rock pile in the dump truck. This results in the water from the nozzle directly impacting the surface of the rock pile, causing water to splash and disperse, making it difficult to directly penetrate the gaps between the rocks, leading to low water resource utilization. Furthermore, it is difficult to effectively remove the clay attached to the deep layers of the rock pile, resulting in cleaning dead zones and difficulty in effectively controlling the mud content of the rock pile.

[0104] Next, referring to the accompanying drawings, a vehicle-mounted large-diameter rockfill water-saving cleaning method according to an embodiment of this application is described.

[0105] Figure 8 This is a flowchart of a water-saving cleaning method for vehicle-mounted large-diameter rockfill provided according to an embodiment of this application.

[0106] like Figure 8 As shown, the vehicle-mounted large-diameter rockfill water-saving cleaning method includes the following steps: In step S801, actual images of the piled stones inside the carriage from multiple perspectives are acquired.

[0107] In step S802, the gap area between the rocks is located based on the actual images from multiple perspectives, and the three-dimensional position information of the gap area is obtained.

[0108] In step S803, a cleaning task for a three-dimensional rinsing operation surface is generated based on the three-dimensional position information of the gap area, so as to execute the cleaning task.

[0109] It should be noted that the foregoing explanation of the vehicle-mounted large-diameter rockfill water-saving cleaning system embodiment also applies to the vehicle-mounted large-diameter rockfill water-saving cleaning method of this embodiment, and will not be repeated here.

[0110] The vehicle-mounted large-diameter rockfill water-saving cleaning method proposed in this application can penetrate deep into the rockfill body through penetrating cleaning. In addition, by acquiring actual images from multiple perspectives, it provides comprehensive and three-dimensional visual data support for gap identification. Furthermore, by accurately locating gap areas and obtaining their three-dimensional position information, it achieves accurate perception of rockfill gaps, providing reliable location basis for cleaning. At the same time, by performing a three-dimensional flushing operation, it abandons flooding cleaning and achieves targeted and precise flushing, which effectively eliminates blind spots in rockfill cleaning, significantly improves the cleanliness of deep rockfill, meets strict mud content control standards, reduces ineffective water consumption from the source, greatly improves water resource utilization, reduces long-term operating costs, and upgrades the cleaning operation from manual and extensive to intelligent and precise through full-process automated control, significantly improving operation efficiency. This solves the problems in related technologies where the nozzle angle is fixed and cannot be flexibly adjusted according to the load and shape of the rock pile in the dump truck. This results in the water from the nozzle directly impacting the surface of the rock pile, causing water to splash and disperse, making it difficult to directly penetrate the gaps between the rocks, leading to low water resource utilization. Furthermore, it is difficult to effectively remove the clay from the deep layers of the rock pile, resulting in cleaning dead zones and difficulty in effectively controlling the mud content of the rock pile.

[0111] Figure 9 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. The electronic device may include: The memory 901, the processor 902, and the computer program stored on the memory 901 and capable of running on the processor 902.

[0112] When the processor 902 executes the program, it implements the vehicle-mounted large-diameter rockfill water-saving cleaning method provided in the above embodiments.

[0113] Furthermore, electronic devices also include: Communication interface 903 is used for communication between memory 901 and processor 902.

[0114] The memory 901 is used to store computer programs that can run on the processor 902.

[0115] The memory 901 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0116] If the memory 901, processor 902, and communication interface 903 are implemented independently, then the communication interface 903, memory 901, and processor 902 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 9 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0117] Optionally, in a specific implementation, if the memory 901, processor 902, and communication interface 903 are integrated on a single chip, then the memory 901, processor 902, and communication interface 903 can communicate with each other through an internal interface.

[0118] The processor 902 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0119] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described vehicle-mounted large-diameter rockfill water-saving cleaning method.

[0120] This application also provides a computer program product, including a computer program that, when executed, implements the above-described vehicle-mounted large-diameter rockfill water-saving cleaning method.

[0121] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0122] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0123] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0124] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium could be paper or other suitable media that can be read by optical scanning, since the program can be obtained electronically by optically scanning the paper or other medium, then editing, interpreting, or otherwise processing it as necessary, and then storing it in a computer memory.

[0125] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0126] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0127] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0128] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A vehicle-mounted water-saving cleaning system for large-diameter rockfill, characterized in that, include: The nozzle array unit is used to perform penetrating cleaning of the target deep rock pile in the riprap of the carriage; An image acquisition unit is used to acquire actual images of the rock pile from multiple perspectives. The gap recognition unit is used to locate the gap region between the piles of stones based on the actual image and obtain the three-dimensional position information of the centroid in the gap region; The central control unit is used to generate a cleaning task for the three-dimensional flushing operation surface based on the three-dimensional position information, so as to drive the nozzle array unit to perform the cleaning task.

2. The system according to claim 1, characterized in that, Also includes: The nozzle angle control unit generates a target angle command for the nozzle array unit according to the cleaning task, so as to drive multiple nozzles of the nozzle array unit to rotate to the corresponding target angle.

3. The system according to claim 1, characterized in that, The nozzle array unit further includes: A horizontal water pipe rack is installed at a predetermined height on the roof of the carriage, with multiple nozzles distributed on the water pipe rack; A vertical water pipe rack and a lifting drive mechanism, wherein the lifting drive mechanism adjusts the vertical height of the vertical water pipe rack relative to the carriage.

4. The system according to claim 1, characterized in that, The gap recognition unit is specifically used to preprocess and enhance the features of the actual image to obtain a processed image that meets preset conditions; identify the pile area in the processed image to generate a pile binary map, and extract the potential gap area in the processed image to generate a gap binary map; perform multi-view image adaptation processing on the pile binary map and the gap binary map according to the acquisition type of the actual image, and perform morphological optimization processing on the gap binary map to determine the spatial relationship between the pile outline and the potential gap outline; Based on the spatial relationship, the gap region is determined to obtain the three-dimensional position information.

5. The system according to claim 4, characterized in that, The gap recognition unit includes: A sub-unit is generated to obtain the perspective transformation matrix based on the corner feature points of the top opening of the carriage. A sub-unit is established to establish the mapping relationship between the image pixel coordinate system and the actual carriage coordinate system based on the top surface of the rubble pile, according to the perspective transformation matrix; The calculation subunit is used to calculate the pixel coordinates of the centroid based on the mapping relationship and the gap region, and convert the pixel coordinates into two-dimensional coordinates of the carriage, so as to generate the three-dimensional position information by combining the distance coordinate value determined based on the average distance of the top or side of the riprap body.

6. The system according to claim 5, characterized in that, The conversion formula for the two-dimensional coordinates of the carriage is: ， ， in, For the horizontal component in homogeneous coordinates. The vertical component in the homogeneous coordinates. The scaling factor in the homogeneous coordinates. This is the perspective transformation matrix. The horizontal coordinate in the pixel coordinates of the centroid. The vertical coordinate in the pixel coordinates of the centroid. The horizontal coordinate in the two-dimensional coordinate system of the carriage is [the horizontal coordinate]. The vertical coordinate in the two-dimensional coordinate system of the carriage is denoted as .

7. A water-saving cleaning method for vehicle-mounted large-diameter rockfill, characterized in that, Includes the following steps: Collect actual images of the piled stones inside the carriage from multiple perspectives; Based on the actual image, locate the gap area between the piles of stones and obtain the three-dimensional position information of the gap area; A cleaning task for a three-dimensional rinsing operation surface is generated based on the three-dimensional position information of the gap region, and the cleaning task is executed.

8. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the vehicle-mounted large-diameter rockfill water-saving cleaning method as described in claim 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the vehicle-mounted large-diameter rockfill water-saving cleaning method as described in claim 7.

10. A computer program product, comprising a computer program, characterized in that, The computer program is executed to implement the vehicle-mounted large-diameter rockfill water-saving cleaning method as described in claim 7.

Images