Underwater hull cleaning robot capable of achieving autonomous detection and method
By designing an autonomously detectable underwater hull cleaning robot, using robotic arms, propeller thrusters and cavitation jet nozzles, and combining neural network models for automatic cleaning, the existing underwater hull cleaning methods are solved, and the problem of low efficiency and narrow application scope of existing underwater hull cleaning methods is solved. , achieving efficient and flexible cleaning results.
Patent Information
- Application Number
- CN202411039674.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-07-31
AI Technical Summary
The existing underwater hull cleaning methods are inefficient, high labor costs, incomplete cleaning, harsh cleaning conditions, and inconvenient cleaning, making it difficult to adapt to complex and harsh underwater environments.
A self-detectable underwater hull cleaning robot is designed, using robotic arms, propeller thrusters, cleaning modules, cameras and control modules, combined with electromagnetic adsorption device and cavitation jet nozzles, image preprocessing and attachment through neural network model Debrief identification, realizing automatic cleaning.
It realizes efficient and flexible cleaning of complex hull surfaces, has a wide range of applications, and can be inspected and cleaned when ships are moored and voyage, improving cleaning efficiency and safety, and reducing the need for manual guidance.
Smart Images

Figure CN118907337A_ABST
Abstract
Description
Technical Field
[0001] The present invention specifically relates to an underwater hull cleaning robot capable of autonomous detection and a method thereof. Background Art
[0002] At present, with the continuous development of domestic technology, most of the current domestic underwater hull cleaning methods have changed from single dock cleaning to more advanced manual underwater cleaning, large cleaning device cleaning or manually controlled underwater robot cleaning. The development of these cleaning methods is more adaptable to the complex situations and harsh environments of underwater operations, and the work efficiency has also been improved. However, these cleaning methods still have problems such as low efficiency, high labor costs, incomplete cleaning, harsh cleaning conditions, and inconvenient cleaning. Therefore, in order to solve the above problems, the present invention provides an underwater hull cleaning robot with high efficiency, comprehensive cleaning and autonomous detection. Summary of the invention
[0003] The purpose of the present invention is to provide an underwater hull cleaning robot and method that can detect autonomously, can flexibly adsorb and climb on the hull, can be applied to the uneven hull surface, and ensure the smooth progress of the cleaning process; it has high flexibility; there is no absolute requirement for the environment in which the hull is located, thereby increasing the scope of application, and during long voyages, it can also check whether important components are adsorbed and attached and clean them in time.
[0004] The technical solution adopted by the present invention is: An underwater hull cleaning robot capable of autonomous detection comprises a frame, and at least two mechanical arms, a propeller thruster, a cleaning module, a camera and a control module arranged on the frame. An electromagnetic adsorption device is arranged at the end of the mechanical arm. A control drive module is connected with the cleaning module, the camera, the mechanical arm, the propeller thruster and the electromagnetic adsorption device. The cleaning module comprises a cavitation jet nozzle.
[0005] Preferably, the frame includes two side panels and a bottom panel, the two side panels are arranged side by side, two ends of the bottom panel are respectively connected to the inner sides of the two side panels, and a support rod is connected between the two side panels. Preferably, a cylindrical cabin is provided on the bottom plate, and the control module is arranged in the cylindrical cabin.
[0006] Preferably, there are two robotic arms, which are arranged on both sides of the frame respectively.
[0007] Preferably, the number of propeller thrusters is 3 to 5.
[0008] Preferably, the cleaning module further comprises a vacuum pump and a water pipe, the cavitation jet nozzle is arranged on the frame, and the vacuum pump is connected to the cavitation jet nozzle through the water pipe.
[0009] Preferably, the control module includes a main controller and a slave controller. The main controller is arranged on the water surface and connected to the slave controller through a cable. The slave controller is arranged in a cylindrical cabin on the rack. The main controller is the upper computer and the slave controller is the lower computer. The slave controller is connected to the robotic arm, propeller thruster, cleaning module and camera.
[0010] Preferably, the rack is also provided with lighting equipment and an optical vision computer, and the optical vision computer is connected to the camera.
[0011] A method for cleaning hull attachments using the above-described autonomously detectable underwater hull cleaning robot, wherein the underwater hull cleaning robot is moved to one side of an underwater hull to be cleaned area, and a camera is used to collect images of underwater hull attachments in the area to be cleaned; Through the trained neural network model based on the fusion attention mechanism and U-Net architecture, the underwater hull images collected by the robot are preprocessed and the features of the underwater hull attachments are extracted. After analysis and processing, the location of the underwater hull attachments is identified; The attachments on the underwater hull are cleaned gradually from the root of the edge of the attachments through the cavitation jet nozzle of the cleaning module.
[0012] Preferably, the specific process of preprocessing the underwater hull images collected by the robot is: graying and enhancing the underwater pictures collected by the robot, segmenting the images through digital image processing detection technology, using threshold-based methods, edge detection segmentation methods, and morphological processing using the Otsu algorithm to segment the images, filter out noise, and extract image information. The beneficial effects of the present invention are: The present invention can flexibly adsorb and climb on the hull through the mechanical arm and the electromagnetic adsorption device at the end thereof, and can be suitable for uneven hull surfaces, better stabilize the underwater hull cleaning robot, and ensure the smooth progress of the cleaning process; the attachments on the hull are cleaned through the cavitation jet nozzle of the cleaning module, which has high flexibility; the underwater hull cleaning robot can clean when the ship is moored through the propeller thruster, and has no absolute requirements for the environment in which the hull is located, thereby increasing the scope of application; during long voyages, it can also check whether important components are adsorbed by attachments and clean them in time; the detection and cleaning work can be carried out autonomously, and only simple manual guidance is required, and the degree of intelligence is relatively good. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 is a three-dimensional diagram of a robot in an embodiment of the present invention; Figure 2 is a cross-sectional view of a robot in an embodiment of the present invention; Figure 3 is a stereoscopic diagram of an underwater hull cleaning robot capable of autonomous detection according to an embodiment of the present invention; Figure 4 It is a flow chart of a method for cleaning hull attachments using the autonomously detectable underwater hull cleaning robot in an embodiment of the present invention; Figure 5 It is a schematic diagram of the movement process of the underwater hull cleaning robot in an embodiment of the present invention.
[0014] In the figure: 1.1-electromagnetic adsorption device, 1.2-mechanical arm; 2.1-frame, 2.2-cavitation jet nozzle, 2.3-cross bar, 2.4-strut, 2.5-cylindrical cabin, 2.6-connecting rod, 2.7-bottom plate, 2.8-propeller thruster; 3.1-vacuum pump, 3.2-water pipe, 3.3-robot. DETAILED DESCRIPTION
[0015] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0016] In the description of the present invention, it should be understood that if the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like are involved, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.
[0017] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection. It can be a mechanical connection or an electrical connection. It can be directly connected or indirectly connected through an intermediate medium. It can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0018] Example 1 An underwater hull cleaning robot capable of autonomous detection, such as Figures 1 to 3 As shown, it includes a frame 2.1, and at least two mechanical arms 1.2, a propeller thruster 2.8, a cleaning module, a camera and a control module arranged on the frame 2.1. An electromagnetic adsorption device 1.1 is provided at the end of the mechanical arm 1.2, and a control drive module is connected with the camera, the mechanical arm 1.2, the propeller thruster 2.8 and the electromagnetic adsorption device 1.1.
[0019] Furthermore, the frame 2.1 comprises two side panels and a bottom panel 2.7. The two side panels are arranged side by side, and two ends of the bottom panel 2.7 are respectively connected to the inner sides of the two side panels. A support rod is also connected between the two side panels. Furthermore, the support rod 2.4 is a round rod.
[0020] Furthermore, a cylindrical cabin 2.5 is provided on the bottom plate 2.7, and the control module is arranged in the cylindrical cabin 2.5.
[0021] A plurality of connecting rods 2.6 are provided below the bottom plate 2.7, and two ends of the connecting rods 2.6 are respectively connected to the two side plates.
[0022] Furthermore, a cross bar 2.3 is connected between the two side plates, and the cavitation jet nozzle 2.2 is arranged on the cross bar 2.3. Furthermore, there are two robotic arms 1.2, which are arranged on both sides of the frame 2.1 respectively.
[0023] Furthermore, the mechanical arms 1.2 symmetrically installed on both sides serve as an adsorption module, which is used to symmetrically install the mechanical arms 1.2 on both sides of the frame 2.1, and install an electromagnetic adsorption device 1.1 with controllable adsorption force on the mechanical arms 1.2 on the surface of the hull.
[0024] Furthermore, the number of propeller thrusters 2.8 is 3 to 5.
[0025] Furthermore, the cleaning module includes a cavitation jet nozzle 2.2, a vacuum pump 3.1 and a water pipe 3.2. The cavitation jet nozzle 2.2 is arranged on the cross bar 2.3 on the frame 2.1. The vacuum pump 3.1 is connected to the cavitation jet nozzle 2.2 through the water pipe 3.2. The cleaning module is used to clean the place to be cleaned by using the cavitation jet principle.
[0026] Furthermore, the control module includes a main controller and a slave controller. The main controller is arranged on the water surface and connected to the slave controller through a cable. The slave controller is arranged in the cylindrical cabin 2.5 on the rack 2.1. The main controller is the upper computer and the slave controller is the lower computer. The slave controller is connected to the robotic arm 1.2, the propeller thruster 2.8, the cleaning module, and the camera.
[0027] A control module, used to control the adsorption module, the driving module, the motion module, the detection module and the cleaning module through a single chip microcomputer; The master-slave control form of the host computer and the slave computer is the control method currently commonly adopted by ROV. The master controller is the host computer, and its main function is to control the motion of the thruster, and it also includes emergency processing, logical operations and other tasks. The slave controller is the slave computer, and its basic responsibility is to execute the instructions issued by the host computer and feedback the status of underwater thrusters and other equipment. The cable maintains the connection between the host computer and the slave computer. The control system mainly controls the posture, motion path and actuator of the ROV in the water, including the surface support part, the control body, the data acquisition device, the communication device, the power management module and the propulsion system module. In view of the unknown nature of the marine environment, ROV needs to have a reliable system that can cope with multiple situations. The present invention uses a computer as the host computer, which mainly realizes the functions of command sending, data receiving, data storage and early warning. The host computer is set as the server end, and the robot is connected by a cable. The thruster is controlled by input devices such as a handle or a main controller, and the output devices such as a touch screen and a display on the shore are used to grasp the motion state of the ROV in real time, and adjust the motion of the thruster at any time in order to achieve the best operating state of the underwater robot.
[0028] As a lower computer, the underwater robot mainly realizes data collection, data storage and data transmission functions. After receiving the command from the upper computer, it changes the speed and direction through the preset program. At the same time, the camera transmits the video data to the upper computer to achieve the purpose of real-time monitoring.
[0029] Furthermore, the rack 2.1 is also provided with lighting equipment and an optical vision computer, and the optical vision computer is connected to the camera.
[0030] Example 2 A method for cleaning hull attachments using the above-mentioned autonomously detectable underwater hull cleaning robot, such as Figures 4 and 5 As shown, the underwater hull cleaning robot is moved to one side of the underwater hull to be cleaned area, and the underwater hull attachments in the area to be cleaned are captured by a camera; Through the trained neural network model based on the fusion attention mechanism and U-Net architecture, the underwater hull images collected by the robot are preprocessed and the features of underwater hull attachments are extracted. After analysis and processing, the location of the attachments is determined. After the location is determined, the robot adjusts its posture through the adsorption robot arm 1.2 to align the cavitation jet nozzle with the edge root of the attachment. The attachments on the underwater hull are cleaned starting from the edge root by the cavitation jet nozzle 2.2 of the cleaning module.
[0031] Furthermore, the specific process of preprocessing the underwater hull images collected by the robot is as follows: grayscale and image enhancement are performed on the underwater pictures collected by the robot, and image segmentation is performed using digital image processing detection technology. The image is segmented based on the threshold method and edge detection segmentation method, and morphological processing uses the Otsu algorithm to segment the image to filter out noise and extract image information; image enhancement is performed on the underwater pictures collected by the robot, and combined with threshold segmentation to further improve the recognition ability of barnacles, etc., and image grayscale, image enhancement linear transformation, and morphological processing are performed. The memory occupied by the collected underwater images is reduced by image grayscale, and the computing efficiency is improved; image enhancement can improve image clarity and contrast; morphological processing uses the Otsu algorithm to segment the image to filter out noise and extract image information.
[0032] Connection method: The main body of the device is connected by rods through two identical racks 2.1; four propeller thrusters 2.8 for controlling the forward and backward movement of the machine are respectively installed on the racks 2.1; the adsorption modules are symmetrically arranged outside the two racks 2.1; a plate is fixed by rods inside the racks 2.1, and a control module and a detection module are arranged on the top, and a propeller thruster 2.8 for controlling the left and right movement of the machine is installed under the rack 2.1; the nozzle of the cavitation jet device is fixed above the control module by another rod, and the rear hose is connected to the pump body.
[0033] Image acquisition and processing Devices required for image acquisition: underwater cameras, lighting equipment, optical vision computers (for underwater image acquisition, image enhancement and recognition), control computers (for decision-making and planning based on observation results, and controlling the robot's next action) Taking hull attachments (mainly barnacles) as the research object, an automatic recognition technology based on deep learning is proposed. This technology constructs a neural network model that integrates the attention mechanism and the U-Net architecture. By capturing the subtle features of the underwater hull attachment images collected by the robot, the collected images are first specially enhanced for the edge of the barnacles, and then the attachments of different sizes and shapes are processed and analyzed. Compared with the existing full convolutional network and U-Net model, this scheme has significantly improved the recognition accuracy. It can effectively improve the detection capability of hull attachments waiting to be detected.
[0034] (I) Dataset A data set was established to measure the criteria for distinguishing attachments. The test data, i.e. photos of hull attachments in underwater environments, were obtained through the Internet and real-life collection. To ensure data quality, images with too much noise or irregular shapes were first excluded, and finally a data set containing multiple environments was formed, each containing 300 images. All images were adjusted to a uniform resolution so that the model could be compatible and trained efficiently.
[0035] The overall dataset is augmented and then randomly shuffled to reduce potential bias or sequence effects. Then, it is divided in a ratio of 6:2:2. This ratio has been widely adopted in the practice of dataset division in many fields because it balances the needs of model training and the verification of generalization ability. Specifically, 60% of the data is used as a training set to learn model parameters; 20% of the data is used as a validation set for adjusting and optimizing hyperparameters in the process to avoid overfitting of the model; the remaining 20% is divided into a test set to evaluate the generalization ability of the model on unknown data, thereby simulating performance in actual application scenarios. Despite the meticulous approach, it should be noted that the dataset comes from a public resource library and may have inherent biases.
[0036] (II) Enhance extraction capabilities based on deep learning and other methods The deep neural network structure is used to automatically extract complex features in the image, and the deep feature extractor is used to identify details such as the edge, width, position and brightness of barnacles that are difficult to capture directly in the image below the waterline. The deep learning algorithm is used for automatic recognition of hull attachments, and the multi-scale deep convolution feature fusion improves the efficiency of hull attachment feature extraction. The supervised learning method adjusts the sample ratio to deal with the problem of data imbalance. In addition, by adding residual connections to the barnacle model on the hull, the ability to extract barnacle features is enhanced.
[0037] After the extraction is completed, the accuracy of the constructed model, that is, the trained neural network model, is judged based on the accuracy rate.
[0038] For classification and recognition performance, accuracy is used as the most direct metric, which is defined as the ratio of correctly classified attachment samples to the total size of the data set. In addition, the confusion matrix is used to calculate TP, TN, FP, and FN to understand the classification performance of the model in more detail. The confusion matrix is a 2x2 matrix in which rows represent actual categories and columns represent predicted categories. According to the definition of the confusion matrix: TP (True Positive): The number of samples correctly identified as positive.
[0039] TN (True Negative): The number of samples correctly identified as negative classes.
[0040] FP (False Positive): The number of samples that are mistakenly identified as positive.
[0041] FN (False Negative): The number of samples incorrectly identified as negative classes.
[0042] Accuracy formula
[0043] By calculating these indicators, the performance of the model can be evaluated more comprehensively, thereby determining the accuracy of the information extracted from the hull attachments. When the accuracy reaches above 0.9, the model training is considered complete.
[0044] 3. Attachment detection method based on feature extraction The method based on feature extraction uses traditional digital image processing technology to pre-process the underwater hull images collected by the robot and extract the features of barnacles and other attachments, and analyze the processed results to determine the location of the barnacles. Digital image processing detection technology image segmentation is based on threshold method, edge detection and other segmentation methods, as well as mathematical morphology methods. Image enhancement is performed on the underwater pictures collected by the robot, combined with threshold segmentation to further improve the recognition ability of barnacles, etc., and several processes such as image graying, image enhancement linear transformation, and morphological processing are performed. Image graying reduces the memory occupied by the collected underwater images and improves computing efficiency; image enhancement can improve image clarity and contrast; morphological processing uses the Otsu algorithm to segment the image, filter out noise, and extract image information.
[0045] Grayscale the images collected by the robot: The underwater hull cleaning robot collects colored images of hull attachments. The images contain various information, and the weighted average method is used to grayscale the images of hull attachments.
[0046] Assuming that the collected crack image is Image(i,j), the gray value of the image VGray(ij) is
[0047] Where: R(ij), G(i,j), B(i,j) are the R, G, BR of the color image point (ij) respectively, G represents the red, and B represents the blue value.
[0048] Image enhancement was performed on the images collected by the robot to improve the image quality. Linear grayscale transformation was used for image enhancement. Morphological processing was performed by using the Otsu algorithm to calculate the maximum inter-class variance, select the threshold, segment the collected barnacle images, extract the barnacle information, and use the morphological principle to filter out the noise and highlight the information of the barnacle image.
[0049] Feature extraction and feature selection are to find the most effective features (invariance of similar samples, discrimination of different samples, robustness to noise) from the original features. The main feature extraction and discrimination: The shell is conical, the shell mouth is serrated, the surface is smooth, and there are no longitudinal ribs. The shell mouth is slightly pentagonal. After the above main features are detected, they are extracted from the image. Feature description or feature vectorization is performed.
[0050] Feature extraction: Convert the original features into a set of features with obvious physical meaning (Gabor, geometric features [corner points, invariants], texture [LBP HOG]) or statistical meaning or kernel. After feature extraction, import it into the discriminant model after deep learning. When the accuracy is greater than 95%, the extraction method is feasible.
[0051] The workflow of the present invention is as follows: pre-manually collect pictures of barnacles and organisms waiting to be cleaned as samples, use a computer to analyze the collected images to obtain analysis data as sample data (sample image), and then perform the following steps: first, select the hull part to be cleaned, and use an underwater camera to collect images of the hull part to be cleaned with the assistance of lighting equipment, and then transmit the collected image information to an optical vision computer for image enhancement (enhanced based on a convolutional neural network, which can rely on modern AI technology) to improve image clarity, etc. in order to better observe the marine organisms to be cleaned, and perform image recognition comparison between the enhanced results and the sample data (image) to determine the category and location information of the organisms to be cleaned, and transmit this information to a control computer for decision analysis (whether cleaning is required, if not, return to the first step and select the next area for operation), control the body to perform cleaning activities on the part to be tested, and finally, repeat the above process. Operation mode: During operation, the underwater components to be cleaned are imported into the model (i.e., the hull model is imported); the map model is constructed using the grid method, and the camera collects real-time information; the algorithm is used to plan the path; the control system controls the robot to run along the planned path, such as Figure 5 shown.
[0052] The sensor collects posture information and transmits it to the main control. The host computer sends the path information to the main control. The main control controls the speed of the propeller thruster 2.8 by controlling the PWM output, thereby realizing the robot's balance, sinking, floating, forward and backward, and left and right turns.
[0053] The camera controls the mechanical arm 1.2 to move to the hull according to the current position, and turns on the electromagnetic adsorption device 1.1 to adsorb to the hull. The camera controls the direction of the cavitation jet nozzle by analyzing the position of the barnacles to achieve the purpose of cleaning the barnacles.
[0054] The hull information is imported during modeling, and the operating area is converted into interconnected but non-overlapping grid units according to a certain division method. The robot's moving path is simplified by converting the complex bottom environment information into discrete grid information.
[0055] In summary, 1. The underwater hull cleaning robot of the present invention has high flexibility and has obvious advantages when cleaning complex structures such as propeller propellers 2.8. 2. The underwater hull cleaning robot of the present invention adopts a method of separating the cavitation jet nozzle from the pump body, which reduces the weight required for the robot structure, reduces energy consumption, improves the flexibility of the robot, and effectively ensures that the robot can operate flexibly in the water. 3. The underwater hull cleaning robot of the present invention adopts electromagnetic adsorption technology. Compared with other adsorption methods, this adsorption method can be applied to uneven hull surfaces, better stabilize the underwater hull cleaning robot, and ensure the smooth progress of the cleaning process. 4. The underwater hull cleaning robot of the present invention can be cleaned when the ship is moored, and there is no absolute requirement for the environment in which the hull is located, which increases the scope of application. During long voyages, it can also check whether important components (such as propellers) are adsorbed by shellfish barnacles and clean them in time. 5. The underwater hull cleaning robot of the present invention has higher practicality. Compared with the commonly used wall-climbing robot cleaning, the underwater hull cleaning robot of the present invention has higher cleaning safety and efficiency. It can clean key parts in a targeted manner, and the time used is shorter. Parts that do not need to be cleaned can be directly skipped. 6. The underwater hull cleaning robot of the present invention can perform detection and cleaning work autonomously, and only requires simple manual guidance, and has a high degree of intelligence.
[0056] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0057] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.
Claims
1. An underwater hull cleaning robot capable of autonomous detection, characterized in that: The invention comprises a frame, and at least two mechanical arms, a propeller propeller, a cleaning module, a camera and a control module arranged on the frame. An electromagnetic adsorption device is arranged at the end of the mechanical arm. The control driving module is connected with the cleaning module, the camera, the mechanical arm, the propeller propeller and the electromagnetic adsorption device. The cleaning module comprises a cavitation jet nozzle.
2. The autonomous underwater hull cleaning robot according to claim 1, characterized in that: The frame comprises two side plates and a bottom plate. The two side plates are arranged side by side. Two ends of the bottom plate are respectively connected to the inner sides of the two side plates. A support rod is also connected between the two side plates.
3. The autonomous underwater hull cleaning robot according to claim 2, characterized in that: A cylindrical cabin is arranged on the bottom plate, and the control module is arranged in the cylindrical cabin.
4. The autonomous underwater hull cleaning robot according to claim 1, characterized in that: There are two robotic arms, which are arranged on both sides of the frame.
5. The autonomously detectable underwater hull cleaning robot according to claim 1, characterized in that: The number of propeller thrusters is 3~5.
6. The autonomously detectable underwater hull cleaning robot according to claim 1, characterized in that: The cleaning module also includes a vacuum pump and a water pipe. The cavitation jet nozzle is arranged on the frame, and the vacuum pump is connected with the cavitation jet nozzle through the water pipe.
7. The autonomously detectable underwater hull cleaning robot according to claim 1, characterized in that: The control module includes a main controller and a slave controller. The main controller is set on the water surface and connected to the slave controller through cables. The slave controller is set on the rack. The main controller is the upper computer and the slave controller is the lower computer. The slave controller is connected to the robotic arm, propeller thruster, cleaning module and camera.
8. The autonomously detectable underwater hull cleaning robot according to claim 1, characterized in that: The rack is also provided with lighting equipment and an optical vision computer, which is connected with the camera.
9. A method for cleaning hull attachments using the autonomously detectable underwater hull cleaning robot as claimed in claim 1, characterized in that: Manipulate the underwater hull cleaning robot to move to one side of the underwater hull to be cleaned area, and collect images of underwater hull attachments in the area to be cleaned through a camera; Through the trained neural network model, the underwater hull images collected by the robot are preprocessed and the features of the underwater hull attachments are extracted. After analysis and processing, the location of the attachments is obtained. After determining the location, the robot adjusts its posture through the adsorption mechanical arm to align the cavitation jet nozzle with the edge root of the attachment. The attachments on the underwater hull are cleaned starting from the root of the edge through the cavitation jet nozzle of the cleaning module.
10. The method for cleaning hull attachments according to claim 9, characterized in that: The specific process of preprocessing the underwater hull images collected by the robot is as follows: grayscale and image enhancement are performed on the underwater pictures collected by the robot, and the images are segmented through digital image processing detection technology, and the threshold method, edge detection segmentation method, and morphological processing method are used to segment the images to filter out noise and extract image information.
Citation Information
Patent Citations
Ocean attachment cleaning plant based on pulsed jet
CN106314717A
Underwater robot with mechanical arms
CN109969363A
SAR image berthing ship target detection method
CN114219996A
Ship cleaning and maintaining device and cleaning and maintaining method thereof
CN115783163A
Industrial part defect detection method, device and equipment and storage medium
CN118014959A
Cited By
Negative pressure adsorption type small-diameter pipeline detection robot and use method
CN119878981A
