CN118083094A

CN118083094A - Tubular underwater operation and maintenance robot system based on shore-machine cooperation and control method

Info

Publication number: CN118083094A
Application number: CN202410271130.2A
Authority: CN
Inventors: 刘军彤; 杜欣; 徐婧雯; 丁宇彤; 甘进; 黄朝炎
Original assignee: Wuhan University of Technology WUT
Current assignee: Wuhan University of Technology WUT
Priority date: 2024-03-11
Filing date: 2024-03-11
Publication date: 2024-05-28

Abstract

The invention discloses a tubular underwater operation and maintenance robot system based on shore-machine cooperation and a control method thereof, wherein the system comprises a shore end control system and an underwater robot system; the shore end control system comprises a carrying ship, and an energy supply module, an operation control module and a signal transmission module which are arranged on the carrying ship, the underwater robot system comprises a detection platform, a cleaning platform and a longitudinal lifting mechanism, the detection platform and the cleaning platform comprise a carrier frame structure, a transverse rotating mechanism and an underwater operation module, and a carrier frame of the detection platform is connected with a carrier frame of the cleaning platform through the longitudinal lifting mechanism; the underwater robot system realizes the omnibearing cleaning and detection of the tubular object under the control of the shore end control system, and simultaneously returns the operation information of the robot in real time to finish the intelligent detection and identification of the positioning and attitude of the robot and the apparent diseases of the underwater structure, and the accurate positioning and analysis. The invention ensures the safety of operation and maintenance operation and improves the accuracy and efficiency of defect detection.

Description

Tubular underwater operation and maintenance robot system based on shore-machine cooperation and control method

Technical Field

The invention particularly relates to a tubular underwater operation and maintenance robot system based on shore-machine cooperation and a control method.

Background

The underwater structure is in water environment for a long time, is influenced by external environmental factors such as water flow, waves, microorganisms in water and the like, is easy to generate various structural diseases, influences the health and safety of the infrastructure, and is easy to cause serious accidents when serious.

Taking a jacket platform as an example, operation and maintenance work of a jacket underwater structure at home and abroad is a main means for divers and underwater robots. The diving operation technology has high maturity and flexible operation, has certain advantages in the operation and maintenance of an offshore shallow water platform, but has low working efficiency, high operation risk and larger environmental influence; compared with manual work, the underwater robot has the advantages that the efficiency is improved, partial working blind areas can be reached, but the operation flexibility and the stability are poor, and the condition that the ROV cannot be recovered due to the winding of the umbilical cable is easy to occur.

Currently, underwater detection techniques are generally divided into two categories: one is based on optical equipment, after relevant images are acquired underwater by utilizing an underwater camera and the like, the acquired images are processed by methods such as image processing and the like to obtain high-quality clear images, and then defects are identified by human eyes or a machine learning detection algorithm; the other is based on acoustic detection equipment, relevant data of the underwater structure are collected through acoustic instruments such as three-dimensional sonar, underwater side-scan sonar and the like, and the health state of the underwater structure is further evaluated through analysis of the data. Generally, the two detection methods are mostly combined, and the two detection methods are combined with each other, so that the two detection methods are indispensable because the two detection methods are combined with each other by utilizing acoustic equipment to perform "area census" and further utilizing optical equipment to perform "local detailed examination". However, the underwater environment is complex and changeable, and when an underwater camera is used for collecting disease images, the images collected underwater are generally low in contrast and unclear due to attenuation of light rays in water and influence of various tiny particles and microorganisms. The characteristics of the underwater image directly lead to difficult feature extraction in the follow-up disease identification, and seriously affect the accuracy and efficiency of disease identification.

Therefore, a tubular underwater operation and maintenance robot system based on shore-machine cooperation and a control method thereof are needed, and meanwhile, an intelligent detection and identification, accurate positioning and analysis method for apparent diseases of underwater structures are needed to be established.

Disclosure of Invention

The invention aims to provide a tubular underwater operation and maintenance robot system based on shore-machine cooperation and a control method thereof, which realize the crawling, cleaning and detection work of a robot along an underwater structure of a tubular, ensure the safety of personnel and greatly improve the control performance and the operation efficiency of the underwater robot. Meanwhile, aiming at the underwater optical detection technology, an intelligent detection and identification method, accurate positioning and analysis method for apparent diseases of the underwater structure are established.

The technical scheme adopted by the invention is as follows:

A tubular underwater operation and maintenance robot system based on shore-machine cooperation comprises a shore end control system and an underwater robot system; the shore end control system comprises a carrying ship, and an energy supply module, a manipulation control module and a signal transmission module which are arranged on the carrying ship, wherein the underwater robot system comprises a detection platform, a cleaning platform and a longitudinal lifting mechanism, the detection platform and the cleaning platform comprise a carrier frame structure, a transverse rotating mechanism and an underwater operation module, both ends of the carrier frame structure are connected with mechanical clamping mechanisms, the transverse rotating mechanism and the underwater operation module are respectively arranged on the outer side and the inner side of the carrier frame structure, a carrier frame of the detection platform is connected with a carrier frame of the cleaning platform through the longitudinal lifting mechanism, the underwater operation module of the detection platform is an underwater detection module, and the underwater operation module of the cleaning platform is an underwater cleaning module;

An electronic cabin module and a navigation positioning module are further arranged on the carrier frame on the detection platform or the cleaning platform, and the electronic cabin module is connected with the transverse rotating mechanism, the longitudinal lifting mechanism, the mechanical clamping mechanism and the navigation positioning module; the control module is connected with the electronic cabin module through the signal transmission module, and the electronic cabin module is connected with the shore end control system through an umbilical cable; the underwater robot system realizes the omnibearing cleaning and detection of the underwater structure of the tubular object under the control of the shore end control system.

Preferably, the manipulation control module comprises a manipulation handle and a control box system, and the manipulation handle is connected with the control box system and is used for sending a manipulation instruction; the control box system is connected with the electronic cabin module through the signal transmission module, receives the instruction and converts the instruction into a control signal to be transmitted to the underwater robot system;

the energy supply module comprises a generator set, a power management system and a power transmission system; the generator sets are driven by fuel oil, and the number of the generator sets is selected according to the operation requirement; one end of the power management system is connected with the generator set, and the other end of the power management system is connected with the control box system and is responsible for monitoring and managing the supply and use conditions of energy.

Preferably, the mechanical clamping mechanism comprises a mechanical arm and an electric push rod, the mechanical arm is connected with the end part of the carrier frame structure through a group of plane four-bar mechanisms, the two ends of the electric push rod are respectively connected with the plane four-bar mechanisms and the carrier frame structure, and a pressure sensor is arranged on the clamping surface of the mechanical arm;

The plane four-bar mechanism comprises a driving piece rocker, a driven piece rocker, a connecting rod and a frame, wherein the frame is fixed at the end part of a carrier frame structure, the end part of the carrier frame structure is hinged with a corresponding mechanical arm through the driven piece rocker, one end of the driving piece rocker is hinged with the carrier frame structure, the other end of the driving piece rocker is hinged with one end of the connecting rod, the other end of the connecting rod is hinged with the mechanical arm, one end of an electric push rod is connected with the driving piece rocker, and the other end of the electric push rod is connected with the carrier frame structure.

Preferably, the transverse rotation mechanism comprises a propeller, a guide wheel and an angle sensor, wherein the propeller is arranged on the outer side of the carrier frame structure, the guide wheel is arranged on the inner side of the carrier frame structure, and the angle sensor is arranged at the right center of the carrier frame structure.

Preferably, the underwater detection module comprises a searchlight, a detection camera, a pressure sensor and a detection cabin, wherein the detection cabin is arranged on the inner side of a carrier frame structure of the detection platform, and the searchlight, the detection camera and the pressure sensor are all arranged on the detection cabin; the front end of the detection cabin is provided with a pressure-resistant cabin of a transparent acrylic semi-sphere cover, and the detection camera is positioned in the pressure-resistant cabin; the position coordinates of the detection camera can be obtained through the pressure sensor and the angle sensor, the depth coordinates h1 of the detection camera are obtained through sensing the water depth through the pressure sensor, meanwhile, the position of the underwater robot, which is tightly held by the tubular pile leg, is taken as an angle zero point before the underwater robot is launched, the angle coordinates theta 1 of the detection camera in the horizontal circumference are obtained through sensing the rotation angle relative to the angle zero point through the angle sensor, the position coordinates (h 1, theta 1) of the detection camera are further obtained, the position coordinates (h 2, theta 2) of the center of the camera are determined through the position of the detection camera, the position (h 2, theta 2) of the center of the detection camera is just seen by the coordinates of the visual focus (the image center) of the underwater structure of the tubular pile, the underwater high-precision measurement algorithm is called, the position correction value (% h, [ theta ]) of the disease relative to the visual focus is measured, and finally the actual position (h 3, theta 3) of the disease is obtained, as shown in fig. 17.

Preferably, the underwater cleaning module comprises a high-pressure water jet device, a cleaning brush, a searchlight, a cleaning camera and a cleaning cabin, wherein the cleaning cabin is arranged on the inner side of a carrier frame structure of the cleaning platform, the high-pressure water jet device, the searchlight and the cleaning camera are arranged on the front surface of the cleaning cabin, and the cleaning brush is arranged on the side surface of the cleaning cabin through a fixed mechanical arm.

Preferably, the longitudinal lifting mechanism comprises a lifter, a motor reducer and a distance alarm, wherein the lifter comprises gear lifters, rack lifting rods and a transmission shaft, each rack lifting rod is sequentially provided with two gear lifters, the gear lifters can move up and down along the rack lifting rods, the two gear lifters are respectively connected and fixed with a carrier frame structure of the detection platform and the cleaning platform, the gear lifter on each rack lifting rod is correspondingly provided with the motor reducer, the motor reducer is connected with the corresponding gear lifter through the transmission shaft, and one end of each rack lifting rod is fixedly connected with the carrier frame structure of the detection platform or the cleaning platform; the detection platform or the cleaning platform is provided with a distance alarm.

Further, the number of the gear lifting rods is two, the gear lifting rods are arranged side by side, the gear lifting machines above the two gear lifting rods are connected through the transmission shaft, the gear lifting machines below the two gear lifting rods are connected through the transmission shaft, and the motor speed reducer is arranged on the transmission shaft above and the transmission shaft below.

Further, the motor speed reducer comprises a servo motor and a worm gear speed reducer, and the servo motor is connected with the transmission shaft through the worm gear speed reducer.

The underwater catheter cleaning method based on the shore-machine cooperation tubular underwater operation and maintenance robot system comprises the following steps of: the detection platform and the cleaning platform are respectively an upper platform and a lower platform;

step S1, operating personnel drive a fully equipped carrying ship to navigate to an underwater structure of a tubular object to be detected;

S2, an operator contracts a longitudinal lifting mechanism of the underwater robot to a minimum stroke, so that the detection platform and the cleaning platform are positioned at adjacent positions; then the underwater robot is moved from the carrying ship to the underwater structure of the tubular object and the installation is completed;

step S3, the mechanical clamping mechanism of the lower platform of the underwater robot is loosened, and the longitudinal lifting mechanism drives the lower platform to vertically move downwards for a certain distance along the tubular pile leg, and then the mechanical clamping mechanism of the lower platform holds the tubular pile leg tightly;

S4, opening an underwater cleaning module and an underwater detection module to clean and detect the surface of the underwater structure of the tubular object; meanwhile, the robot is driven by the transverse rotating mechanism to do circumferential motion in the horizontal plane around the underwater structure of the tubular object, so that 360-degree omnibearing operation is realized;

Step S5, during the rotation process of the robot, an operator observes the cleaning effect of the surface of the underwater structure of the tubular object through a cleaning camera arranged above the underwater cleaning module; if the cleaning effect is poor, repeating the step S4, and simultaneously starting a cleaning brush in the underwater cleaning module to clean the residual attachments for the second time; if the cleaning effect is good and the disease detection result is not affected, closing the underwater cleaning module and performing step S6;

S6, loosening a mechanical clamping mechanism of an upper platform of the underwater robot, and tightly holding the tubular pile leg by the mechanical clamping mechanism of the upper platform after the longitudinal lifting mechanism drives the upper platform to vertically move downwards for a certain distance along the tubular pile leg;

step S7, repeating the steps S3-S6 until the lower platform meets a complex truss node or other obstacles of the underwater structure of the tubular object or the bottom of the underwater structure of the tubular object, and jumping out of circulation;

the jump-out cycle when the lower platform encounters a truss node or other obstacle includes the following steps:

S8, contracting a longitudinal lifting mechanism of the underwater robot to a minimum stroke to enable an upper platform and a lower platform of the underwater robot to be in adjacent positions, and enabling the lower platform to abut against truss nodes or obstacles;

step S9, the underwater robot loosens the mechanical clamping mechanism of the lower platform, the longitudinal lifting mechanism drives the lower platform to vertically move downwards along the tubular pile leg to avoid truss nodes or obstacles by a certain distance, and then the mechanical clamping mechanism of the lower platform holds the tubular pile leg tightly;

Step S10, similarly, the mechanical clamping mechanism of the upper platform of the underwater robot is loosened, the longitudinal lifting mechanism drives the upper platform to vertically downwards pass through the node along the tubular pile leg, and then the mechanical clamping mechanism of the upper platform holds the tubular pile leg tightly;

Step S11, repeating the steps S3-S7 until the pipe moves to the bottom of the underwater structure of the pipe and jumps out of circulation;

The method comprises the following steps after encountering a bottom jump-out cycle of a tubular underwater structure: the underwater robot loosens the mechanical clamping mechanisms of the upper platform and the lower platform and falls off from the pile leg of the tubular object, and an operator drags the underwater robot to the sea level through the traction rope to complete the recovery work of the underwater robot and carry out the detection work of the next pile leg.

Preferably, in the step S5, in the process of the transverse rotation of the robot, the underwater detection module performs real-time video recording and automatically identifies disease information on the surface of the underwater structure of the tubular, marks the extracted pictures containing the disease and establishes files, specifically including disease types, levels, positions and the like, and uniformly stores the files in a designated folder for later evaluation of the health state of the underwater structure of the tubular.

As shown in fig. 16, the specific process of the "underwater detection module for real-time video recording and automatically identifying disease information on the surface of the underwater structure of the tubular object, marking the extracted photos containing the disease and establishing files, and uniformly storing in the designated folder" is as follows:

s5.1, intercepting an image acquired in real time by an underwater detection module according to the number of frames;

S5.2, carrying out sharpening treatment on each frame of image by adopting technologies such as data enhancement, image fusion and the like;

Step S5.3, calling a trained 'Yolo V-based high-precision detection and identification network of the surface defects of the underwater structure';

Step S5.4, detecting whether each frame image contains defects through an identification network, if the frame image does not have defects, skipping the frame image, detecting the next frame image, and if the frame image has defects, extracting the images containing the defects;

S5.5, calibrating the rough outline of the defect through the difference of pixel values in the image;

step S5.6, calling a standard data set (library) of the surface defects of the underwater structure;

S5.7, comparing the type and the grade of the defect again;

s5.8, calling the position coordinates of diseases in the frame of image;

step S5.9, renaming the image containing the defect to be 'disease type-level-position';

and S5.10, uniformly storing the images containing the defects in a designated folder.

In the step 5.3, the training process of the 'Yolo V-based high-precision detection and identification network of the surface defects of the underwater structure' is as follows:

Step 5.3.1, reading image data in the underwater structure surface disease data set;

Step 5.3.2, dividing the read image into a plurality of Patch files;

step 5.3.3, preprocessing the Patch file based on the data enhancement and image fusion technology;

Step 5.3.4, randomly dividing the preprocessed image into a training set (train) and a test set (test) according to a certain proportion (the proportion is 4:1) (one part is used as the training set for training the detection model, the other part is used as the test set for verifying the training effect), respectively storing the training set and the test set in two corresponding folders train and test, and repeatedly training the target detection model YOLOv through a training file; different parameters in a training file can be modified to control the training of the model, and when the detection precision (the detection precision is the precision rate and the recall rate) of the model reaches more than 90%, the target detection model is stored to obtain the high-precision detection and identification network for the surface defects of the underwater structure based on Yolov.

In the step 5.3.3, the "preprocessing the Patch file based on the data enhancement and image fusion technology" mainly comprises white balance, pyramid fusion and histogram equalization, and the specific process is as follows:

Step 5.3.3.1, performing white balance treatment on the underwater image to obtain a color-corrected image Input 1;

step 5.3.3.2, performing CLAHE algorithm processing and bilateral filtering processing on the image Input1 to obtain an image Input 2;

Further, in step 5.3.3.2, the specific process of performing the CLAHE algorithm processing and the bilateral filtering processing on the image Input1 to obtain the image Input 2 is as follows: aiming at the problems of low contrast, more noise and the like of the image Input1 after color correction, firstly, the L component in the Lab space is processed by a CLAHE algorithm to enhance the contrast of the image; carrying out bilateral filtering treatment on the enhanced image, reducing noise in the image and enhancing image details; obtaining an image Input 2 after the twice processing;

Step 5.3.3.3, respectively calculating four weight graphs of the preprocessed images Input1 and Input 2: global contrast weight map W _C, local contrast weight map W _LC, chromaticity weight map W _S, saliency weight map W _E;

step 5.3.3.4, carrying out normalization processing on the four weight graphs to obtain a normalized weight graph;

Further, in step 5.3.3.4, the specific process of normalizing the four weight graphs is:

The calculation formula of the normalization process is as follows:

In the method, in the process of the invention, -Normalized weights of the a-th image;

-the sum of the four weights of the a-th image; the value range of a is 1-2, and the value ranges respectively represent images Input1 and Input2;

step 5.3.3.5, carrying out Laplacian pyramid decomposition on the two Input images Input 1 and Input 2 to obtain Laplacian pyramid images, and carrying out Gaussian pyramid decomposition on the normalized weight graph to obtain standard weight Gaussian pyramid images, wherein the number of layers of the pyramid images is 5;

step 5.3.3.6, carrying out fusion processing on the Laplacian pyramid image of the input image and the Gaussian pyramid image corresponding to the standard weight on each layer to obtain a fused pyramid image F;

The first layer image calculation formula of the image F is as follows:

wherein/> For the pyramid image F after the fusion process,Normalized weights for layer I image of Gaussian pyramid image,/>Normalized weights for the first layer image of the laplacian pyramid image;

Step 5.3.3.7, starting from the top layer, upsampling the pyramid image F, namely performing interpolation expansion operation on the first layer image F _l to make the size of the first layer image equal to that of the first-1 image, and adding the expanded image F _l and F _l-1 to obtain a new image of the first-1 layer; sequentially operating from top to bottom to finally obtain an image with the same size as the input image, namely finishing the output image after the sharpening process; (l represents the number of layers of the image, in this embodiment, the value range of l is 1-5, that is, "the number of layers of the pyramid image is 5 layers" in step 5.3.3.5. Generally, the number of layers of the pyramid is related to the size of the image and the detail information that the image needs to retain: 1) the size of the image: the number of layers of the pyramid is generally influenced by the size of the original image, and for large-size images, more layers of pyramids may be required for fusion processing in order to better capture detailed information; 2) Detail information that the image needs to retain: if more detail features are required to be reserved in the image fusion, the number of layers of the pyramid is correspondingly increased so as to facilitate better processing and expressing the detail features of the image; the number of layers of the pyramid is generally determined according to a specific usage scenario to achieve the best effect. )

The calculation formula of the sharpening process is as follows:

In the method, in the process of the invention, -Outputting an image; /(I)A new image generated for the kth time in the top-down accumulation process;

Up-sampling.

For example: the surface disease data set of the underwater structure totally comprises 1000 disease images with the size of 256 pixels multiplied by 256 pixels, wherein 600 images are collected in experimental pools under different turbidity conditions, and 400 images are generated by DCGAN networks. The data set mainly comprises a plurality of categories of slightly, moderately, severely blurred and slightly gridded generated images; according to the requirements of network training and model verification, randomly dividing a data set into a training set (train) and a test set (test) according to the ratio of 4:1;

in the target detection and recognition network YOLOv, the training parameters are as follows:

1) Learning rate (LEARNING RATE): the learning rate determines the update amplitude of the training model in each iteration, a larger learning rate may lead to an unstable training process, and a smaller learning rate may lead to too slow training speed or a fall into a locally optimal solution;

2) Batch Size (Batch Size): the batch size determines the number of samples used for each update of the model. The larger batch size can increase training speed, but also increases memory and computing resource requirements;

3) Iteration number (Number of Epochs): the number of iterations determines the number of rounds of model training. Too few iterations may result in a model under-fit, while too many iterations may result in a model over-fit;

4) Loss Function (Loss Function): the loss function defines the difference between the model prediction and the real label, and different loss functions can influence the learning effect of the model on targets in different categories and positions;

The choice of these training parameters above is generally dependent on the specific use scenario and data set, e.g. the design of the loss function can be adjusted for different number of target classes; parameters such as batch size and learning rate can be adjusted for different hardware resources. Therefore, the selection of training parameters generally needs to be adjusted in combination with the actual situation to obtain the best performance; the values of the training parameters in this embodiment are: the learning rate was 0.0002, the batch size was 64, and the number of iterations was 2000.

It should be noted that the underwater structure surface disease data set used in the model training contains various types of diseases such as cracks, breakage, etc. and different levels of the same type of diseases, so that the call of "Yolo v-based underwater structure surface defect high-precision detection and identification network" can judge whether a certain frame of image contains a disease or not, and initially obtain the type and level of the disease, and the higher the level is, the more serious the disease is represented.

The beneficial effects of the invention are as follows:

1. The underwater robot system realizes the omnibearing cleaning and detection of the underwater structure of the tubular object under the control of the shore end control system, and simultaneously transmits the operation information of the underwater robot back to the PC end upper computer system in real time through the signal transmission module, thereby completing the intelligent detection and identification of the positioning and attitude of the underwater robot and the apparent diseases of the underwater structure, and the accurate positioning and analysis.

2. The invention realizes the intelligent detection and identification of the apparent diseases of the underwater structure, improves the precision and efficiency of defect detection, and realizes the comprehensive analysis of the diseases of the underwater structure, thereby acquiring the health condition of the underwater structure, and having important significance for guaranteeing the health and safety of the underwater structure of the traffic infrastructure; meanwhile, the invention realizes the crawling, cleaning and detecting work of the robot along the underwater structure of the tubular object, so that the robot has the capability of automatically crossing complex truss nodes and obstacles of the tubular object, greatly increases the movement flexibility of the robot, replaces manual operation, avoids the danger of manual operation, can realize underwater unmanned operation, can complete the cleaning and detecting of the underwater structure of the tubular object by simply operating on a carrying ship, reduces the operation difficulty of operators and ensures the safety of personnel; in addition, the information closed loop is formed by establishing the information interaction model of the underwater robot system and the shore end control system, so that the control accuracy and the operation efficiency of the underwater robot are greatly improved; the invention is especially suitable for the underwater structure of the jacket and is also suitable for all similar tubular wading structures.

Drawings

FIG. 1 is a diagram of an overall system architecture of a tubular underwater operation and maintenance robot system based on shore-machine cooperation in an embodiment of the present invention;

FIG. 2 is a schematic diagram of an operating state of a tubular underwater operation and maintenance robot system based on shore-machine cooperation in an embodiment of the present invention;

FIG. 3 is a schematic diagram of information interaction of a tubular underwater operation and maintenance robot system based on shore-machine cooperation in an embodiment of the present invention;

FIG. 4 is a schematic diagram of a shore end control system according to an embodiment of the present invention;

FIG. 5 is a schematic view of a carrying vessel according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a configuration of a submersible robot system according to an embodiment of the present invention;

FIG. 7 is a schematic view of an electronic module according to an embodiment of the present invention;

FIG. 8 is a schematic diagram of a motion control module according to an embodiment of the present invention;

FIG. 9 is a schematic diagram of a structure of a underwater detection module according to an embodiment of the present invention;

FIG. 10 is a schematic view of a submerged cleaning module according to an embodiment of the present invention;

FIG. 11 is a flowchart of a control method of a tubular underwater operation and maintenance robot system based on shore-machine cooperation in an embodiment of the present invention;

FIG. 12 is a node obstacle avoidance flow chart of the underwater robot system in an embodiment of the present invention;

FIG. 13 is a flow chart of motion control of the underwater robot system according to the embodiment of the present invention;

FIG. 14 is a flow chart of the underwater cleaning of the underwater robot system in the embodiment of the present invention;

FIG. 15 is a flow chart of the underwater detection of the underwater robot system according to the embodiment of the present invention;

FIG. 16 is a flowchart of an automatic underwater disease identification algorithm of the underwater robot system in the embodiment of the present invention;

FIG. 17 is a flowchart of an underwater disease precise positioning algorithm of the underwater robot system in an embodiment of the present invention;

FIG. 18 is a flowchart of an intelligent detection and identification method for surface diseases of underwater structures of the underwater robot system according to the embodiment of the present invention;

In the figure: 1a shore end control system and 2 an underwater robot system;

1-1 of a carrying ship, 1-1-1 of a ship body, 1-1-1 of a ship deck, 1-1-1-2 of a propulsion system, 1-1-2 of a robot storage area, 1-1-3 of a generator set storage area, and 1-1-4 of a personnel operation area;

1-2 energy supply modules, 1-2-1 generator sets, 1-2-2 power management systems, 1-2-3 power transmission systems;

1-3 control modules, 1-3-1 control handles, 1-3-2PC end upper computer systems and 1-3-3 control box systems;

1-4 signal transmission modules;

a 2-1 electronics module, a 2-1-1 communication system, a 2-1-2 control system, a 2-1-3 sensor system;

2-2 navigation positioning module, 2-2-1 gesture sensing system, 2-2-2GPS positioning system;

2-3 a motion control module;

2-3-1 mechanical clamping mechanism, 2-3-1-1 mechanical arm, 2-3-1-2 electric push rod, 2-3-1-3 pressure sensor;

a 2-3-2 longitudinal lifting mechanism, a 2-3-2-1 lifter, a 2-3-2-2 motor reducer and a 2-3-2-3 distance alarm;

2-3-3 transverse rotating mechanism, 2-3-3-1 guide wheel, 2-3-3-2 propeller and 2-3-3-3 angle sensor;

2-4 underwater operation modules;

2-4-1 underwater cleaning module, 2-4-1-1 cleaning water gun, 2-4-1-2 cleaning brush, 2-4-1-3 searchlight, 2-4-1-4 cleaning camera;

the device comprises a 2-4-2 underwater detection module, a 2-4-2-1 searchlight, a 2-4-2-2 detection camera, a 2-4-2-3 angle sensor and a 2-4-2-4 pressure sensor.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

In the description of the present invention, it should be understood that if there are terms such as "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc., the indicated azimuth or positional relationship is based on the azimuth or positional relationship shown in the drawings, it is merely for convenience of description and simplification of the description, and does not indicate or imply that the indicated apparatus or element must have a specific azimuth, be constructed and operated in a specific azimuth, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.

In the description of the present invention, it should be noted that the terms "mounted," "connected," and "coupled" are to be construed broadly, as well as, for example, fixedly coupled, detachably coupled, or integrally coupled, unless otherwise specifically indicated and defined. Either mechanically or electrically. Can be directly connected or indirectly connected through an intermediate medium, and can be communicated with the inside of two elements or the interaction relationship of the two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.

Example 1

A tubular underwater operation and maintenance robot system based on shore-machine cooperation is shown in figures 1-10, and comprises a shore end control system 1 and an underwater robot system 2; the shore end control system 1 comprises a carrying ship 1-1, and an energy supply module 1-2, a manipulation control module 1-3 and a signal transmission module 1-4 which are arranged on the carrying ship 1-1, wherein the underwater robot system 2 comprises a detection platform, a cleaning platform and a longitudinal lifting mechanism 2-3-2, the detection platform and the cleaning platform comprise a carrier frame structure, a transverse rotating mechanism 2-3-3 and an underwater operation module 2-4, two ends of the carrier frame structure are connected with a mechanical clamping mechanism 2-3-1, the transverse rotating mechanism 2-3-3 and the underwater operation module 2-4 are respectively arranged on the outer side and the inner side of the carrier frame structure, a carrier frame of the detection platform is connected with a carrier frame of the cleaning platform through the longitudinal lifting mechanism 2-3-2, the underwater operation module of the detection platform is an underwater detection module, and the underwater operation module of the cleaning platform is an underwater cleaning module;

The electronic cabin module 2-1 and the navigation positioning module 2-2 are further arranged on the carrier frame on the detection platform or the cleaning platform, and the electronic cabin module 2-1 is connected with the transverse rotating mechanism 2-3-3, the longitudinal lifting mechanism 2-3-2, the mechanical clamping mechanism 2-3-1 and the navigation positioning module 2-2; the control module 1-3 is connected with the electronic cabin module 2-1 through the signal transmission module 1-4, and the electronic cabin module 2-1 is connected with the control module 1-3 of the shore end control system 1 through an umbilical cable; the underwater robot system 2 realizes the omnibearing cleaning and detection of the underwater structure of the tubular object under the control of the shore end control system 1; the mechanical clamping mechanism 2-3-1, the transverse rotating mechanism 2-3-3 and the longitudinal lifting mechanism 2-3-2 together form a motion control module 2-3.

Example 2

As shown in fig. 1 to 5, the parts of the shore end control system 1 are further defined on the basis of the embodiment 1, and the performance of the defined embodiment 2 is more excellent.

The control module 1-3 comprises a control handle 1-3-1 and a control box system 1-3-3, wherein the control handle 1-3-1 is connected with the control box system 1-3-3 and is used for sending control instructions; the control box system 1-3-3 is connected with the electronic cabin module 2-1 of the underwater robot system 2 through the signal transmission module 1-4, the control box system 1-3-3 receives the instruction and converts the instruction into a control signal to be transmitted to the underwater robot system 2, the signal transmission module 1-4 adopts an optical fiber communication mode to transmit the control signal to the underwater robot system 2, and meanwhile operation information and the like of the underwater robot are transmitted back to the control box system 1-3-3;

the control box system 1-3-3 is connected with the PC end upper computer system 1-3-2 and is used for receiving information such as the position, the posture and the operation image of the underwater robot.

The navigation positioning module 2-2 includes a gesture sensing system 2-2-1 and a GPS positioning system 2-2-2.

The energy supply module 1-2 comprises a generator set 1-2-1, a power management system 1-2-2 and a power transmission system 1-2-3; the generator sets 1-2-1 are driven by fuel oil, and the number of the generator sets is selected according to the operation requirement; one end of the power management system 1-2-2 is connected with the generator set 1-2-1, and the other end is connected with the control box system 1-3-3, and is responsible for monitoring and managing the supply and use conditions of energy sources.

Further, the carrying ship 1-1 is a double-body operation ship, and comprises a ship body 1-1-1, a robot storage area 1-1-2, a generator set 1-2-1 storage area 1-1-3 and a personnel operation area 1-1-4; the ship body 1-1-1 consists of a ship deck 1-1-1-1 and a propulsion system 1-1-1-2; the personnel operation area 1-1-4 and the equipment storage area are both positioned above the ship deck 1-1-1; the carrying ship 1-1 serves as a temporary operation platform on water, and operation and maintenance operations can be conveniently and rapidly developed.

Example 3

As shown in fig. 6 to 10, the parts of the underwater robot system 2 are further defined on the basis of the embodiments 1 and 2, and the performance of the defined embodiment 3 is more excellent.

The mechanical clamping mechanism 2-3-1 comprises a mechanical arm 2-3-1-1 and an electric push rod 2-3-1-2, the mechanical arm 2-3-1-1 is connected with the end part of the carrier frame structure through a group of plane four-bar mechanisms, the two ends of the electric push rod are respectively connected with the plane four-bar mechanisms and the carrier frame structure, and a pressure sensor 2-3-1-3 is arranged on the clamping surface of the mechanical arm 2-3-1-1; the pressure sensor 2-3-1-3 is used for monitoring the clasping force of the mechanical arm 2-3-1-1 encircling the tubular object, and the mechanical clamping mechanism 2-3-1 stops working when the clasping force is greater than the set value 1;

The plane four-bar mechanism comprises a driving piece rocker, a driven piece rocker, a connecting rod and a frame, wherein the frame is fixed at the end part of a carrier frame structure, the end part of the carrier frame structure is hinged with a corresponding mechanical arm 2-3-1-1 through the driven piece rocker, one end of the driving piece rocker is hinged with the carrier frame structure, the other end of the driving piece rocker is hinged with one end of the connecting rod, the other end of the connecting rod is hinged with the mechanical arm 2-3-1-1, one end of an electric push rod 2-3-1-2 is connected with the driving piece rocker, and the other end of the electric push rod 2-3-1-2 is connected with the carrier frame structure.

Further, the transverse rotating mechanism 2-3-3 comprises a propeller 2-3-3-2, a guide wheel 2-3-3-1 and an angle sensor 2-3-3-3, wherein the propeller 2-3-3-2 is arranged on the outer side of the carrier frame structure, the guide wheel 2-3-3-1 is arranged on the inner side of the carrier frame structure, and the angle sensor 2-3-3-3 is arranged at the center of the carrier frame structure; the angle sensor 2-3-3-3 is used for monitoring the transverse rotation angle of the underwater robot, so as to realize the pose and positioning of the robot.

Further, the underwater detection module 2-4-2 comprises a searchlight 2-4-2-1, a detection camera 2-4-2-2, an angle sensor 2-4-2-3, a pressure sensor 2-4-2-4 and a detection cabin, wherein the detection cabin is arranged on the inner side of a carrier frame structure of the detection platform, and the searchlight 2-4-2-1, the detection camera 2-4-2-2, the angle sensor 2-4-2-3 and the pressure sensor 2-4-2-4 are all arranged on the detection cabin; the front end of the detection cabin is provided with a pressure-resistant cabin of a transparent acrylic semi-sphere cover, and the detection camera is positioned in the pressure-resistant cabin; the detection camera 2-4-2-2 is used for shooting disease information on the surface of the tubular object, and accurate positioning of diseases is realized by means of various sensors.

The specific thinking that the detection camera is used for shooting disease information on the surface of a tubular object and realizing accurate positioning of diseases by means of various sensors is as follows: to realize accurate positioning of diseases, the position coordinates of the detection camera need to be determined, when the robot system is designed, the angle sensor and the pressure sensor are tightly close to the detection camera, so the position coordinates of the detection camera can be accurately obtained through the two types of sensors (the specific process of accurately obtaining the position coordinates of the detection camera through the two types of sensors is that the pressure sensor is used for measuring water pressure, and then the depth coordinates of the detection camera, namely the longitudinal coordinate Z, are obtained; then the specific position of the disease is determined by detecting the picture shot by the camera, as shown in detail in fig. 17.

Further, the underwater cleaning module 2-4-1 comprises a high-pressure water jet device, a cleaning brush 2-4-1-2, a searchlight 2-4-1-3, a cleaning camera 2-4-1-4 and a cleaning cabin, wherein the cleaning cabin is arranged on the inner side of a carrier frame structure of the cleaning platform, the high-pressure water jet device, the searchlight 2-4-1-3 and the cleaning camera 2-4-1-4 are arranged on the front surface of the cleaning cabin, and the cleaning brush 2-4-1-2 is arranged on the side surface of the cleaning cabin through a fixed mechanical arm.

Further, the high-pressure water jet equipment is a cleaning water gun 2-4-1-1, and the cleaning camera 2-4-1-4 is used for monitoring the cleaning effect of the surface of the tubular object and assisting the cleaning equipment to perform fixed-point cleaning.

Further, the longitudinal lifting mechanism 2-3-2 comprises a lifting machine 2-3-2-1, a motor reducer 2-3-2-2 and a distance alarm 2-3-2-3, wherein the lifting machine 2-3-2-1 comprises gear lifting machines, rack lifting rods and transmission shafts, each rack lifting rod is sequentially provided with two gear lifting machines, the gear lifting machines can move up and down along the rack lifting rods, the two gear lifting machines are respectively connected and fixed with a carrier frame structure of the detection platform and a carrier frame structure of the cleaning platform, the gear lifting machines on each rack lifting rod are correspondingly provided with the motor reducer 2-3-2-2, the motor reducer 2-3-2-2 is connected with the corresponding gear lifting machines through the transmission shafts, and one end of each rack lifting rod is fixedly connected with the carrier frame structure of the detection platform or the cleaning platform; the detection platform or the cleaning platform is provided with a distance alarm 2-3-2-3 for monitoring the distance between the underwater robot detection platform and the cleaning platform, and when the distance is greater than a set value 2 or less than a set value 3, the longitudinal lifting mechanism stops moving.

Further, the number of the gear lifting rods is two, the gear lifting rods are arranged side by side, the gear lifting machines above the two gear lifting rods are connected through a transmission shaft, the gear lifting machines below the two gear lifting rods are connected through the transmission shaft, and motor speed reducers 2-3-2-2 are arranged on the transmission shaft above and the transmission shaft below.

Further, the motor speed reducer 2-3-2-2 comprises a servo motor and a worm gear speed reducer, and the servo motor is connected with a transmission shaft through the worm gear speed reducer.

The electronic cabin module 2-1 comprises a control system 2-1-2, a sensor system 2-1-3 and a communication system 2-1-1, wherein the communication system 2-1-1, the sensor system 2-1-3 and the navigation positioning module 2-2 are connected with the control system 2-1-2, and the control system 2-1-2 is connected with a detection platform, a transverse rotating mechanism 2-3-3 of a cleaning platform, an underwater operation module 2-4, a mechanical clamping mechanism 2-3-1 and a longitudinal lifting mechanism 2-3-2 for controlling the work of each part of the robot.

The electronic cabin module 2-1 and the navigation positioning module 2-2 are arranged in a pressure-resistant cabin of the underwater robot system 2, and the pressure-resistant cabin is arranged in a detection cabin of the detection platform and is transparent. The electronic cabin module 2-1 comprises a communication system 2-1-1, a control system 2-1-2 and various sensor systems 2-1-3, and is responsible for information processing and exchange of the shore end control system 1 and the underwater robot system 2; the navigation and positioning module 2-2 comprises an attitude sensor, a GPS system and the like and is used for collecting information of the position, the attitude and the like of the underwater robot. Example 4

The control method of the underwater operation and maintenance robot system for the tubular object based on shore-machine cooperation, as shown in fig. 11-15, comprises the following steps: the detection platform and the cleaning platform are respectively an upper platform and a lower platform;

Step S3, the mechanical clamping mechanism of the lower cleaning platform of the underwater robot is loosened, and the longitudinal lifting mechanism drives the lower platform to vertically move downwards for a certain distance along the tubular pile leg, and then the mechanical clamping mechanism of the lower platform holds the tubular pile leg tightly;

Further, in the step S5, during the transverse rotation of the robot, the underwater detection module performs real-time video recording and automatically identifies the disease information on the surface of the underwater structure of the tubular object, marks the extracted photos containing the disease and establishes files, specifically including disease types, levels, positions and the like, and uniformly stores the files in a designated folder for later evaluation of the health status of the underwater structure of the tubular object.

As shown in fig. 16, the specific idea of "the underwater detection module performs real-time video recording and automatically recognizes disease information on the surface of the tubular object, marks the extracted photos containing the disease and creates files, and uniformly stores the files in a designated folder" is as follows: the method comprises the steps of obtaining a high-quality and clear detection image of an underwater structure through technical means such as data enhancement and image fusion, automatically identifying disease information on the surface of the tubular object by adopting a ' Yolo v-based high-precision detection and identification network model for surface defects of the underwater structure ' (a ' Yolo v-based high-precision detection and identification network model for surface defects of the underwater structure ' is arranged in a detection camera or a control system '), simultaneously extracting a picture of the identified disease, automatically marking the outline of the disease, naming the picture name as a disease type-level-coordinate, and uniformly storing the picture name in a designated folder for evaluating the health condition of the underwater structure.

The specific process of carrying out real-time video recording by the underwater detection module and automatically identifying disease information on the surface of the tubular object, marking the extracted pictures containing the disease and establishing files, and uniformly storing the files in a designated folder is as follows:

S5.7, comparing the type and the grade of the defect again;

s5.8, calling the position coordinates of diseases in the frame of image;

In the step 5.3, the training process of the high-precision detection and identification network of the surface defects of the underwater structure based on Yolo V is as follows:

Step 5.3.2, dividing the read image into a plurality of Patch files;

Step 5.3.4, randomly dividing the preprocessed image into a training set train and a test set test according to a certain proportion (the proportion is 4:1) (one part is used as a training set for training a detection model, the other part is used as a test set for verifying a training effect), respectively storing the training set train and the test set test in two corresponding folders train and test, and repeatedly training a target detection model YOLOv through a training set file; different parameters in the training file can be modified to control the training of the model, and when the detection precision (the detection precision is the precision rate and the recall rate) of the target detection model by the test set file reaches more than 90%, the target detection model is stored to obtain the Yolov-based high-precision detection and identification network for the surface defects of the underwater structure.

In the step 5.3.3, preprocessing is performed on the Patch file based on the data enhancement and image fusion technology, and the preprocessing mainly comprises white balance, pyramid fusion and histogram equalization, and the specific process is as follows:

The calculation formula of the normalization process is as follows:

-the sum of the four weights of the a-th image;

The first layer image calculation formula of the image F is as follows:

step 5.3.3.7, starting from the top layer, upsampling the pyramid image F, namely performing interpolation expansion operation on the first layer image F _l to make the size of the first layer image equal to that of the first-1 image, and adding the expanded image F _l and F _l-1 to obtain a new image of the first-1 layer; sequentially operating from top to bottom to finally obtain an image with the same size as the input image, namely finishing the output image after the sharpening process;

The calculation formula of the sharpening process is as follows:

In the method, in the process of the invention, -Outputting an image;

Up-sampling.

The intelligent detection and identification method for the surface diseases of the underwater structure mainly comprises the steps of clearing the underwater image and automatically identifying the underwater diseases, wherein the underwater image is processed by a data enhancement and image fusion method by means of an autonomously established underwater structure surface disease data set, so that the color of the underwater image is corrected, the foggy and fuzzy degree of the underwater image is reduced, the global contrast of the image is improved, the integrity of the edge of a detection target is ensured, the quality of the underwater image is improved, and the problems of low detection precision and poor detection efficiency caused by poor image quality when the detection is carried out by means of machine vision and the like are effectively solved; through YOLOv target detection algorithm, the high-precision and high-efficiency automatic detection of the surface diseases of the underwater structure is realized, a new thought and a practical strategy are provided for complex detection and safety state evaluation of the surface diseases of the underwater structure, the technical problem of low detection precision of the underwater structure at present is effectively overcome, and the method has wide and good application prospect.

In summary, the invention discloses a tubular underwater operation and maintenance robot system based on shore-machine cooperation and a control method thereof, comprising a shore end control system and an underwater robot system; the shore end control system comprises a carrying ship, and an energy supply module, an operation control module and a signal transmission module which are arranged on the carrying ship, the underwater robot system comprises a detection platform, a cleaning platform and a longitudinal lifting mechanism, the detection platform and the cleaning platform comprise a carrier frame structure, a transverse rotating mechanism and an underwater operation module, two ends of the carrier frame structure are connected with mechanical clamping mechanisms, the transverse rotating mechanism and the underwater operation module are respectively arranged on the outer side and the inner side of the carrier frame structure, and the carrier frame of the detection platform is connected with the carrier frame of the cleaning platform through the longitudinal lifting mechanism; the underwater robot system realizes the omnibearing cleaning and detection of the underwater structure of the tubular object under the control of the shore end control system, and simultaneously transmits the operation information of the underwater robot back to the PC end upper computer system in real time through the signal transmission module, thereby completing the intelligent detection and identification of the positioning and attitude of the underwater robot and the apparent diseases of the underwater structure, and the accurate positioning and analysis.

The invention realizes intelligent detection and identification of apparent diseases of the underwater structure to a certain extent, improves the accuracy and efficiency of defect detection, and realizes comprehensive analysis of the diseases of the underwater structure, thereby obtaining the health condition of the underwater structure, and having important significance for guaranteeing the health and safety of the underwater structure of the traffic infrastructure. The defect detection method is not limited to a tubular underwater structure, but is applicable to all wading structures, including underwater parts of bridges, reservoir dams, ocean platforms, ships and the like, and underwater structures such as submarine tunnels, submarines and the like are also covered. The material of the wading structure is not limited, and reinforced concrete, steel structure or other materials are suitable.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.

In summary, ⑴ is based on the operation idea of shore-machine cooperation, and the information closed loop is formed by establishing the information interaction model of the underwater robot system 2 and the shore-end control system 1, so that the control accuracy and the operation efficiency of the underwater robot are greatly improved. ⑵ The invention provides a control method for the underwater robot to autonomously cross the complex truss node or the obstacle of the tubular structure based on an alternate motion strategy, and the motion flexibility of the robot is greatly improved. ⑶ The invention can realize unmanned operation underwater, and an operator can finish cleaning and detection of an underwater structure only by simply operating on a carrying ship, thereby reducing the operation difficulty of the operator and ensuring the safety of the operator. ⑷ The invention provides a complete cleaning and detecting method for the underwater structure, and simultaneously establishes an intelligent detection and identification, accurate positioning and analysis method for apparent diseases of the underwater structure, improves the accuracy and efficiency of defect detection, and realizes comprehensive analysis of the diseases of the underwater structure, thereby acquiring the health condition of the underwater structure, and having important significance for guaranteeing the health and safety of the underwater structure of a traffic infrastructure.

It will be understood that modifications and variations will be apparent to those skilled in the art from the foregoing description, and it is intended that all such modifications and variations be included within the scope of the following claims.

Claims

1. A tubular object underwater operation and maintenance robot system based on shore-machine cooperation is characterized in that: the system comprises a shore end control system and an underwater robot system; the shore end control system comprises a carrying ship, and an energy supply module, a manipulation control module and a signal transmission module which are arranged on the carrying ship, wherein the underwater robot system comprises a detection platform, a cleaning platform and a longitudinal lifting mechanism, the detection platform and the cleaning platform comprise a carrier frame structure, a transverse rotating mechanism and an underwater operation module, both ends of the carrier frame structure are connected with mechanical clamping mechanisms, the transverse rotating mechanism and the underwater operation module are respectively arranged on the outer side and the inner side of the carrier frame structure, a carrier frame of the detection platform is connected with a carrier frame of the cleaning platform through the longitudinal lifting mechanism, the underwater operation module of the detection platform is an underwater detection module, and the underwater operation module of the cleaning platform is an underwater cleaning module;

An electronic cabin module and a navigation positioning module are further arranged on the carrier frame on the detection platform or the cleaning platform, and the electronic cabin module is connected with the transverse rotating mechanism, the longitudinal lifting mechanism, the mechanical clamping mechanism and the navigation positioning module; the control module is connected with the electronic cabin module through the signal transmission module.

2. The shore-machine collaboration based tubular underwater operation and maintenance robot system of claim 1, wherein: the control module comprises a control handle and a control box system, and the control handle is connected with the control box system and is used for sending control instructions; the control box system is connected with the electronic cabin module through the signal transmission module, receives the instruction and converts the instruction into a control signal to be transmitted to the underwater robot system;

3. The shore-machine collaboration based tubular underwater operation and maintenance robot system of claim 1, wherein: the mechanical clamping mechanism comprises a mechanical arm and an electric push rod, the mechanical arm is connected with the end part of the carrier frame structure through a group of planar four-bar mechanisms, the two ends of the electric push rod are respectively connected with the planar four-bar mechanisms and the carrier frame structure, and a pressure sensor is arranged on the clamping surface of the mechanical arm;

4. The shore-machine collaboration based tubular underwater operation and maintenance robot system of claim 1, wherein: the transverse rotating mechanism comprises a propeller, a guide wheel and an angle sensor, wherein the propeller is arranged on the outer side of the carrier frame structure, the guide wheel is arranged on the inner side of the carrier frame structure, and the angle sensor is arranged at the right center of the carrier frame structure.

5. The shore-machine collaboration based tubular underwater operation and maintenance robot system of claim 1, wherein: the underwater detection module comprises a searchlight, a detection camera, a pressure sensor and a detection cabin, wherein the detection cabin is arranged on the inner side of a carrier frame structure of the detection platform, and the searchlight, the detection camera and the pressure sensor are all arranged on the detection cabin; the front end of the detection cabin is provided with a pressure-resistant cabin containing a transparent hemispherical cover, and the detection camera is positioned in the pressure-resistant cabin;

The underwater cleaning module comprises a high-pressure water jet device, a cleaning brush, a searchlight, a cleaning camera and a cleaning cabin, wherein the cleaning cabin is arranged on the inner side of a carrier frame structure of the cleaning platform, the high-pressure water jet device, the searchlight and the cleaning camera are arranged on the front surface of the cleaning cabin, and the cleaning brush is arranged on the side surface of the cleaning cabin through a fixed mechanical arm.

6. The shore-machine collaboration based tubular underwater operation and maintenance robot system of claim 1, wherein: the longitudinal lifting mechanism comprises lifters, motor reducers and distance alarms, each lifter comprises a gear lifter, a rack lifter and a transmission shaft, each rack lifter is sequentially provided with two gear lifters, each gear lifter can move up and down along the rack lifter, the two gear lifters are respectively connected and fixed with a carrier frame structure of the detection platform and the cleaning platform, each gear lifter on each rack lifter is correspondingly provided with one motor reducer, each motor reducer is connected with the corresponding gear lifter through the transmission shaft, and one end of each rack lifter is fixedly connected with the carrier frame structure of the detection platform or the cleaning platform; the distance alarm is arranged on the detection platform or the cleaning platform.

7. A control method of the underwater operation and maintenance robot system for tubular objects based on shore-machine cooperation according to claim 1, which is characterized in that: the method comprises the following steps: the detection platform and the cleaning platform are respectively an upper platform and a lower platform;

S2, an operator contracts a longitudinal lifting mechanism of the underwater robot to a minimum stroke, so that the detection platform and the cleaning platform are in a close-fitting position; then the underwater robot is moved from the carrying ship to the underwater structure of the tubular object and the installation is completed;

Step S5, during the rotation process of the robot, an operator observes the cleaning effect of the surface of the underwater structure of the tubular object through a cleaning camera arranged above the underwater cleaning module; if the cleaning effect is poor, repeating the step S4, and simultaneously starting a cleaning brush in the underwater cleaning module to clean the residual attachments for the second time; if the cleaning effect is good and the disease detection effect is not affected, closing the underwater cleaning module and performing step S6;

S8, contracting a longitudinal lifting mechanism of the underwater robot to a minimum stroke to enable an upper platform and a lower platform of the underwater robot to be in a close-fitting position, and enabling the lower platform to be close to truss nodes or obstacles;

Step S10, similarly, the mechanical clamping mechanism of the upper platform of the underwater robot is loosened, the longitudinal lifting mechanism drives the upper platform to vertically downwards pass through the node along the pile leg, and then the mechanical clamping mechanism of the upper platform holds the pile leg of the tubular object tightly;

8. The control method according to claim 7, characterized in that: in the step S5, during the transverse rotation of the robot, the underwater detection module performs real-time video recording and automatically identifies the disease information on the surface of the underwater structure of the tubular object, marks the extracted pictures containing the disease and establishes files, specifically including the profile, type, level, position and the like of the disease, and the pictures are uniformly stored in a designated folder for later evaluation of the health state of the underwater structure of the tubular object.

9. The control method according to claim 8, characterized in that: the method comprises the following steps that an underwater detection module carries out real-time video recording and automatically identifies disease information on the surface of a tubular object, marks the extracted pictures containing the disease, establishes files, and uniformly stores the files in a designated folder, and the specific implementation steps are as follows:

S5.3, calling a trained 'high-precision detection and identification network for surface defects of underwater structures';

s5.6, calling an underwater structure surface defect standard data set;

S5.7, comparing the type and the grade of the defect again;

s5.8, calling the position coordinates of diseases in the frame of image;

10. The control method according to claim 9, characterized in that: in the step 5.3, the training process of the 'high-precision detection and identification network of the surface defects of the underwater structure' is as follows:

Step 5.3.2, dividing the read image into a plurality of Patch files;

And 5.3.4, randomly dividing the preprocessed image into a training set train and a test set test according to a certain proportion, respectively storing the training set train and the test set test, repeatedly training the target detection model through a training set file, and storing the target detection model when the detection precision of the target detection model through the test set file reaches more than 90%, so as to obtain the high-precision detection and identification network for the surface defects of the underwater structure.