Intelligent docking device and method for unloading arm

Abstract

The application provides a kind of unloading arm intelligent docking device and intelligent docking method, it is related to intelligent docking technical field, the position information of target flange interface is collected by image recognition module, the motion angle information of unloading arm and ship motion angle information are collected by sensor module, control system is based on the position information, the motion angle information of unloading arm and ship motion angle information collected, the movement path of unloading arm is planned, and corresponding control instruction is generated to control driving module to drive unloading arm to complete docking.The scheme provided in the application realizes the automatic control of the unloading arm, at the same time, improves the accuracy and efficiency of the unloading arm docking, ensures the stable connection of the unloading arm and the target flange interface, thereby simplifying the docking operation process and reducing the labor intensity of the operator.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent docking, in particular to an unloading arm intelligent docking device and an intelligent docking method. BACKGROUND

[0002] The unloading arm is a key equipment connecting the transport ship and the storage and transportation system, and is widely used in the loading and unloading operation of low-temperature or high-pressure medium such as liquefied natural gas, liquefied petroleum gas, ethylene, and liquid ammonia, and is mainly deployed in coastal receiving stations, inland liquefied wharfs, chemical industry parks, and large storage tank areas, for realizing safe, efficient, and airtight material transfer between the ship and the shore.

[0003] In the related art, the traditional docking method of the unloading arm is performed by manual remote control. Since the working place is usually located at a wharf, etc., the information of the target flange interface position needs to be transmitted in real time. However, since the traditional manual docking not only requires skilled operation skills of workers, but also requires timely communication and exchange of flange position information between the ship and the shore workers, and requires investment of certain manpower, due to the complex conditions of the working place, not only the overall docking work is difficult, but also the risk coefficient is increased. SUMMARY

[0004] To solve or partially solve the problems in the related art, the present application provides an unloading arm intelligent docking device and an intelligent docking method, which can effectively solve the problem of difficult manual docking of the unloading arm.

[0005] The first aspect of the present application provides an unloading arm intelligent docking device, which comprises an image recognition module, a sensor module, a control system, and a driving module;

[0006] The image recognition module, the sensor module, and the driving module are installed on the unloading arm;

[0007] The image recognition module is configured to collect target flange interface images and ship body motion angle information in real time;

[0008] The sensor module is configured to collect motion angle information of the unloading arm and ship body motion angle information in real time;

[0009] The control system is configured to monitor the motion of the unloading arm in real time according to the data collected by the image recognition module and the sensor module, and issue a control instruction;

[0010] The driving module is installed at a joint of the unloading arm, and is configured to receive the control instruction and drive the unloading arm to move to the target flange interface based on the control instruction.

[0011] As an embodiment of the first aspect of the present application, the control system is further configured to:

[0012] establishing a discharge arm reference coordinate system based on a base of the discharge arm;

[0013] establishing a D-H coordinate system based on the discharge arm reference coordinate system, for representing the movement direction of each joint of the discharge arm;

[0014] establishing a coordinate system of a three-dimensional joint of the discharge arm based on the discharge arm reference coordinate system, for representing the position of the three-dimensional joint relative to the base;

[0015] constructing an inverse kinematics equation according to the D-H coordinate system and the coordinate system of the three-dimensional joint, and solving the inverse kinematics equation combined with a mathematical twin model to obtain the movement angle of each joint of the discharge arm running to the target flange interface position.

[0016] As an embodiment of the first aspect of the application, the image recognition module is equipped with an intelligent recognition model, which is used to analyze the relative position of the discharge arm and the target flange interface according to the target flange interface image collected by the image recognition module;

[0017] After the image recognition module collects the target flange interface image, it extracts the required image frame by frame and creates a data set;

[0018] After data labeling of the data set, the intelligent recognition model is obtained by training through the YOLO algorithm.

[0019] As an embodiment of the first aspect of the application, the image recognition module includes a monocular camera and a binocular camera;

[0020] The monocular camera is installed on the column of the discharge arm, used to collect the first flange interface position image, and the target flange interface required for docking by the discharge arm is selected through the intelligent recognition model;

[0021] The binocular camera is installed on the three-dimensional joint of the discharge arm, used to collect the second flange interface position image, and the position of the target flange interface required for docking by the discharge arm is accurately judged through the intelligent recognition model;

[0022] The control system completes the preliminary docking of the target flange interface according to the first position information, and completes the secondary docking of the target flange interface according to the second position information.

[0023] As an embodiment of the first aspect of the application, the sensor module includes a pull rope sensor, an inclination sensor, and an ultrasonic sensor;

[0024] The pull rope sensor is installed on the column shaft box of the unloading arm and the three-dimensional joint connection, and is used to collect the horizontal rotation angle of the unloading arm and the horizontal rotation angle of the three-dimensional joint.

[0025] The inclination sensor is installed on the inner arm and the outer arm of the unloading arm, and is used to collect the joint angle of the inner arm and the outer arm in real time.

[0026] The ultrasonic sensor is installed on the three-dimensional joint of the unloading arm, and is used to collect the distance information of the target flange interface from the unloading arm.

[0027] As an embodiment of the first aspect of the application, the control system comprises an unloading arm control system and a motion control system.

[0028] The unloading arm control system is installed on the upper computer, and is used to monitor the running state of the unloading arm and generate a control instruction for controlling the running of the unloading arm.

[0029] The motion control system is installed on the lower computer, and is used to receive the data collected by the image recognition module and the sensor module, and control the driving module according to the control instruction.

[0030] As an embodiment of the first aspect of the application, the control system further comprises a flange interface position prediction system, which is used to simulate the ship body sway caused by sea waves based on the ship body motion angle information collected by the image recognition module, and predict the offset or deformation trend of the flange interface position.

[0031] As an embodiment of the first aspect of the application, the driving module comprises an inner arm driving module, an outer arm driving module and a horizontal driving module.

[0032] The inner arm driving module is used to drive the inner arm joint of the unloading arm to perform vertical motion based on the control instruction.

[0033] The outer arm driving module is used to drive the outer arm joint of the unloading arm to perform vertical motion based on the control instruction.

[0034] The horizontal driving module is used to drive the inner arm joint of the unloading arm to perform horizontal motion based on the control instruction.

[0035] The second aspect of the application provides an intelligent docking method of an unloading arm, which comprises:

[0036] Collecting a first flange interface position image and obtaining first position information of a target flange interface according to the first flange interface position image;

[0037] Control the driving module to drive the unloading arm to complete the preliminary docking according to the first position information;

[0038] Collect a second flange interface position image, and obtain second position information of the target flange interface according to the second flange interface position image;

[0039] Collect distance information of the target flange interface from the unloading arm;

[0040] Control the driving module to drive the unloading arm to complete the secondary docking according to the second position information and the distance information.

[0041] As an embodiment of the second aspect of the application, the control of the driving module to drive the unloading arm to complete the preliminary docking according to the first position information further comprises:

[0042] Real-time collection of the ship body movement angle information for real-time prediction of the target flange interface position, generation of the preliminary docking control instruction based on the prediction result, the first position information and the distance information, and control of the driving module to drive the unloading arm to complete the preliminary docking according to the preliminary docking control instruction;

[0043] The control of the driving module to drive the unloading arm to complete the secondary docking according to the second position information and the distance information further comprises:

[0044] Real-time collection of the ship body movement angle information for real-time prediction of the target flange interface position, generation of the secondary docking control instruction based on the prediction result, the second position information and the distance information, and control of the driving module to drive the unloading arm to complete the secondary docking according to the secondary docking control instruction.

[0045] The technical scheme provided by the application can have the following beneficial effects: the intelligent docking device for the unloading arm is composed of an image recognition module, a sensor module, a control system and a driving module, etc., realizes automatic control of the unloading arm, accurately completes the docking process of the unloading arm, improves the accuracy and efficiency of the docking of the unloading arm, can adjust the docking mode according to different working environments, has strong adaptability. At the same time, through accurate position information recognition, sensor data collection and automatic control, accurate docking of the unloading arm and the target flange interface can be realized, stable connection of the unloading arm and the target flange interface is ensured, liquid leakage problem in the docking process is effectively avoided, safety is improved, and on the basis of automatic control, the operation process is simplified, and the labor intensity of the operator is reduced.

[0046] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the application. BRIEF DESCRIPTION OF DRAWINGS

[0047] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which like reference characters refer to like parts throughout the several views.

[0048] Figure 1 is a schematic diagram of the structure of the unloading arm shown in the embodiments of the present application;

[0049] Figure 2 is a schematic diagram of the architecture of the intelligent docking device of the unloading arm shown in the embodiments of the present application;

[0050] Figure 3 is a schematic diagram of the intelligent docking method flow of the unloading arm shown in the embodiments of the present application.

[0051] Symbol explanation:

[0052] 1 - base, 2 - stand, 3 - inner arm, 4 - outer arm, 5 - docking device, 51 - three-dimensional joint, 52 - three-dimensional joint claw, 53 - guide plate, 54 - manifold, 6 - image recognition module, 61 - monocular camera, 62 - binocular camera, 7 - sensor module, 71 - pull cord sensor, 72 - inclination sensor, 73 - ultrasonic sensor, 8 - control system, 81 - unloading arm control system, 82 - motion control system, 83 - flange interface position prediction system, 9 - drive module, 91 - inner arm drive module, 92 - outer arm drive module, 93 - horizontal drive module. DETAILED DESCRIPTION

[0053] Embodiments of the present application will be described more fully hereinafter with reference to the accompanying drawings, in which embodiments of the application are shown. This application may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.

[0054] The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0055] It should be understood that, although the terms "first", "second", "third", etc. can be used in this application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information without departing from the scope of the application. Therefore, the features defined as "first", "second" can explicitly or implicitly include one or more features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0056] The unloading arm is a key equipment connecting the transport ship and the storage and transportation system, and is used for realizing safe, efficient and closed material transfer between the ship and the shore. In the related art, the traditional docking mode of the unloading arm is realized by manual remote control docking. Since the traditional manual docking not only requires skilled operation skills of workers, but also requires timely communication and exchange of flange position information of workers at the ship and the shore, and requires investment of certain manpower, due to the complex conditions of the working place, not only the overall docking work is difficult, but also the risk coefficient is increased.

[0057] In view of the above problems, the embodiments of the present application provide an unloading arm intelligent docking device and an intelligent docking method. The unloading arm intelligent docking device is composed of an image recognition module, a sensor module, a control system and a driving module, which improves the accuracy and efficiency of the unloading arm docking, and effectively solves the problem of manual docking difficulty of the unloading arm.

[0058] In the embodiments of the present application, an unloading arm intelligent docking device is provided, which comprises an image recognition module 6, a sensor module 7, a control system 8 and a driving module 9.

[0059] The image recognition module 6, the sensor module 7 and the driving module 9 are installed on the unloading arm.

[0060] The image recognition module 6 is used for real-time acquisition of target flange interface image and ship body motion angle information.

[0061] The sensor module 7 is used for real-time acquisition of the motion angle information of the unloading arm and the ship body motion angle information.

[0062] The control system 8 is used for real-time monitoring of the motion of the unloading arm according to the data collected by the image recognition module 6 and the sensor module 7 and issuing control instructions.

[0063] The driving module 9 is installed at the joint of the unloading arm, and the driving module is used for receiving the control instructions and driving the unloading arm to move to the target flange interface based on the control instructions.

[0064] The main structure of the unloading arm includes a base 1, a stand 2, an inner arm 3, an outer arm 4, and a docking device 5, and the specific structure is shown in Figure 1

[0065] The docking device 5 includes a three-dimensional joint 51, a three-dimensional joint claw 52, a guide plate 53, and a manifold 54.

[0066] The three-dimensional joint 51 is used to provide free movement of the docking device 5 in three-dimensional space.

[0067] The three-dimensional joint claw 52 is used to lock the target flange interface after the unloading bag is docked with the target flange interface.

[0068] The guide plate 53 is used to improve the positioning basis for the movement of the docking device 5.

[0069] The manifold 54 is used to connect the target flange interface.

[0070] The image recognition module 6 and the sensor module 7 are electrically connected with the control system 8, the control system 8 is electrically connected with the driving module 9, the image recognition module 6 and the sensor module 7 send the collected data to the control system 8 for processing, the control system 8 generates control instructions according to the collected data, and displays the state of the unloading arm in real time, the control instructions are sent to the driving module 9 by the control system 8, and the driving module 9 controls each joint of the unloading arm to operate according to the control instructions.

[0071] In the embodiment, based on the image recognition module 6, the sensor module 7, the control system 8, and the driving module 9, the intelligent docking device of the unloading arm is composed, the automatic control of the unloading arm is realized, the docking process of the unloading arm is accurately completed, the accuracy and efficiency of the docking of the unloading arm are improved, the docking mode can be adjusted according to different working environments, and the adaptability is high. At the same time, through accurate position information recognition, sensor data acquisition and automatic control, accurate docking of the unloading arm and the target flange interface can be realized, stable connection of the unloading arm and the target flange interface is ensured, liquid leakage problem in the docking process is effectively avoided, safety is improved, and on the basis of automatic control, the operation process is simplified, and the labor intensity of the operator is reduced.

[0072] In the embodiment of the application, the control system is further used for:

[0073] A reference coordinate system of the unloading arm is established based on the base 1 of the unloading arm.

[0074] A D-H coordinate system is established based on the reference coordinate system of the unloading arm, and is used to represent the movement direction of each joint of the unloading arm.

[0075] ​Wherein, the D-H coordinate system represents the running direction of the inner arm 3 relative to the column 2, the movement direction of the outer arm 4 relative to the inner arm 3, and the movement direction of the docking device 5 relative to the outer arm.

[0076] The coordinate system of the three-dimensional joint 51 of the unloading arm is established based on the unloading arm reference coordinate system, for representing the position of the three-dimensional joint 51 relative to the base.

[0077] The inverse kinematics equation is constructed according to the D-H coordinate system and the coordinate system of the three-dimensional joint 51, and the joint movement angles of the unloading arm running to the target flange interface position are solved by combining the mathematical twin model.

[0078] Wherein, the parameters required in the inverse kinematics analysis can be obtained according to the D-H coordinate system, including the link length a and the joint offset d。

[0079] The inverse kinematics equation is as shown in formula (1).

[0080] Formula (1);

[0081] x, y, z are the axis parameters of the coordinate system of the three-dimensional joint 51, is the length of the inner arm 3, is the length of the outer arm 4, is the offset degree of the outer arm 4-docking device 5 joint, is the offset degree of the inner arm 3-outer arm 4 joint, is the offset degree of the column 2-inner arm 3 joint, is the required movement angle of the docking device 5, is the required movement angle of the outer arm, is the required movement angle of the inner arm.

[0082] In the embodiment, the inverse kinematics analysis can provide an accurate theoretical basis for the movement control of the unloading arm. By constructing the coordinate systems of multiple joints and performing inverse kinematics analysis, the current joint angle and the required movement angle of each joint are obtained, which provides a data basis for accurate planning of the movement path of the unloading arm, thereby realizing accurate control of the unloading arm.

[0083] In the embodiment of the present application, the image recognition module 6 is equipped with an intelligent recognition model, which is used to analyze the relative position of the unloading arm and the target flange interface according to the target flange interface image collected by the image recognition module 6.

[0084] After the image recognition module 6 collects the target flange interface image, the required image is extracted by frame, and a data set is created.

[0085] The number of extracted frames can be set in the control system 8 according to actual needs.

[0086] After data labeling of the data set, the intelligent recognition model is obtained by training through the YOLO algorithm.

[0087] Among them, the roboflow tool is used to label the target flange interface image in the data set, and the labeling parameter is the relative position of the target flange interface in the image.

[0088] In the secondary docking process, when the three-dimensional joint 51 is 40 cm away from the target flange, the target flange interface image collected by the image recognition module 6 is a target flange interface incomplete image, in order to improve the docking accuracy, the fitting circle code is added in the YOLO algorithm, which is used to fit the complete target flange interface according to the position of some points on the target flange interface incomplete image, so as to realize the accurate identification of the target flange interface.

[0089] After the target flange interface is identified, the identification result is verified, and when the matching degree of the predicted box and the real box is greater than 85%, it is considered that the intelligent recognition model training is completed.

[0090] In this embodiment, the intelligent recognition model obtained by using the YOLO algorithm is used to identify the position of the target flange interface, and the fitting circle code is added to complete the target flange interface incomplete image, which greatly improves the intelligent degree and accuracy of image recognition, and provides an accurate position basis for the docking of the discharge arm.

[0091] In the embodiment of the application, the image recognition module 6 includes a monocular camera 61 and a binocular camera 62;

[0092] The monocular camera 61 is installed on the stand column 2 of the discharge arm, used to collect the first flange interface position image, and select the target flange interface required to be docked by the discharge arm through the intelligent recognition model, to obtain the first position information of the target flange interface.

[0093] The binocular camera 62 is installed on the three-dimensional joint 51 of the discharge arm, used to collect the second flange interface position image, and accurately judge the position of the target flange interface required to be docked by the discharge arm through the intelligent recognition model, to obtain the second position information of the target flange interface.

[0094] Among them, the data set created by the image recognition module 6 includes a first flange interface image data set and a second flange interface image data set, the first flange interface image data set is created based on the first flange interface position image collected by the monocular camera 61, and the second flange interface image data set is created based on the second flange interface position image collected by the binocular camera 62.

[0095] The control system 8 completes the preliminary docking of the target flange interface according to the first position information, and completes the secondary docking of the target flange interface according to the second position information.

[0096] In the embodiment, a large amount of image data is collected by the image recognition module 6 to create a data set, and position information is obtained based on analysis of the data set, which significantly improves the accuracy of the obtained position information, enhances the robustness of the algorithm, and enables the image recognition module to adapt to higher-precision collection scenarios. In addition, the preliminary docking based on the obtained first position information enables the unloading arm to quickly approach the target flange interface, improving the speed of intelligent docking, and the secondary docking based on the obtained second position information enables the unloading arm to make accurate adjustments near the target flange interface, greatly improving the docking accuracy. The method based on two dockings enables the unloading arm to adapt to complex environments, reduces error accumulation, and ensures high-quality completion of the docking task.

[0097] In the embodiment of the present application, the sensor module 7 includes a pull rope sensor 71, an inclination sensor 72, and an ultrasonic sensor 73.

[0098] The pull rope sensor 71 is installed at the connection between the column shaft box of the unloading arm and the three-dimensional joint 51, and is used to collect the horizontal rotation angle of the unloading arm and the horizontal rotation angle of the three-dimensional joint 51.

[0099] The inclination sensor 72 is installed on the inner arm 3 and the outer arm 4 of the unloading arm, and is used to collect the joint angles of the inner arm 3 and the outer arm 4 in real time, so as to monitor the angle values moved during the automatic docking process and ensure that the current movement angle is the same as the calculated required angle.

[0100] The ultrasonic sensor 73 is installed on the three-dimensional joint 51 of the unloading arm, and is used to collect distance information of the target flange interface from the unloading arm.

[0101] The data collected by the pull rope sensor 71 is a digital quantity, and the pull rope sensor 71 needs to be calibrated to convert the digital quantity data into angle data. The calibration process includes:

[0102] According to the angle data read by the inclination sensor 72 and the digital quantity data read by the pull rope sensor 71, data curves are respectively generated, the lowest point and the highest point of the data curve of the inclination sensor 72 are selected and recorded, and the maximum value and the minimum value of the digital quantity information of the pull rope sensor 71 are selected and recorded. The maximum value and the minimum value of the two curves are fitted through the fitting curve code in matlab, and the curve fitting is completed.

[0103] According to the numerical value of the P value polynomial generated after the curve fitting is completed, the numerical value of the corresponding position of the two pull rope sensors 71 is changed, so as to complete the calibration of the pull rope sensor 71. After the calibration is completed, the digital quantity data collected by the pull rope sensor 71 corresponds to the angle data, which can be directly converted into angle data.

[0104] Further, in order to ensure the accurate collection of distance information by the ultrasonic sensor 73, an ultrasonic sensor is installed on the top, bottom, left and right of the rear side of the pipe of the three-dimensional joint 51. Through mean value calculation of the distance information collected by the four ultrasonic sensors, accurate distance information of the target flange interface from the unloading arm is obtained.

[0105] In the embodiment, the data collected by the various sensors can provide more comprehensive and accurate environmental perception information for the docking of the unloading arm, make up for the shortcomings of a single sensor, enhance the robustness and reliability, and at the same time, the digital quantity data collected by the pull rope sensor 71 is calibrated by the inclination sensor 72, the accurate conversion of the digital quantity data is realized, and the pull rope sensor 71 can accurately obtain the angle data of multiple joints, so as to provide accurate angle data for the docking of the unloading arm.

[0106] In the embodiment of the present application, the control system 8 includes an unloading arm control system 81 and a motion control system 82.

[0107] The unloading arm control system 81 is installed on the upper computer, which is used to monitor the running state of the unloading arm and generate control instructions for controlling the running of the unloading arm.

[0108] The upper computer includes but is not limited to a computer, a tablet, a mobile device, etc.

[0109] The inverse kinematics analysis is based on the unloading arm control system 81, and the unloading arm control system 81 generates control instructions for controlling the running of the unloading arm based on the inverse kinematics analysis.

[0110] The unloading arm control system 81 is equipped with a visualization module, which displays the running state of the unloading arm in real time according to the received first position information, the first position information and the distance information of the target flange interface from the unloading arm.

[0111] The motion control system 82 is installed on the lower computer, which is used to receive the data collected by the image recognition module 6 and the sensor module 7, and control the driving module 9 according to the control instructions.

[0112] The motion control system 82 feeds back the received data to the unloading arm control system 81 for analysis and processing.

[0113] Specifically, the lower computer is a PLC, and the motion control system 82 is carried in the PLC. The motion control system 82 is electrically connected with the image recognition module 6, the sensor module 7 and the driving module 9 through a cable based on the PLC. The motion control system 82 is electrically connected with the unloading arm control system 81 through a wireless communication mode or a wired communication mode. The specific connection structure is shown in Figure 2

[0114] The wireless communication mode includes but is not limited to Bluetooth, WiFi, zigbee and the like, and the wired communication mode includes but is not limited to a network cable, USB, CAN and the like.

[0115] Further, the unloading arm control system 81 generates a control instruction through path planning. When the unloading arm starts to dock, the monocular camera 61 is first opened to take an image of the target flange interface to obtain a first flange interface image data set. The first position information of the target flange interface is obtained based on intelligent recognition model analysis of the first flange interface image data set. The unloading arm control system 81 analyzes the first position information based on inverse kinematics to plan a path for the movement of the unloading arm to a specified position and generates a preliminary docking control instruction. The motion control system 82 controls the unloading arm to complete preliminary docking along the planned path according to the preliminary docking control instruction.

[0116] Then the binocular camera 62 is opened to take an image of the target flange interface to obtain a second flange interface image data set. The second position information of the target flange interface is obtained based on intelligent recognition model analysis of the second flange interface image data set. The unloading arm control system 81 analyzes the first position information based on inverse kinematics to plan a path for the movement of the unloading arm to a specified position and generates a secondary docking control instruction. The motion control system 82 controls the unloading arm to complete secondary docking along the planned path according to the preliminary docking control instruction. When the unloading arm is too close to the target flange interface, the ultrasonic sensor 73 is used to judge the distance and feedback to the unloading arm control system 81 in real time. The unloading arm control system 81 sends the feedback distance information to the motion control system 82. The motion control system 82 controls the driving device 9 to complete precise secondary docking of the unloading arm according to the distance information.

[0117] In the embodiment, the monocular camera 61, the binocular camera 62 and the sensor cooperate to enable the unloading arm intelligent docking device to accurately position the target, thereby providing accurate data basis for the unloading arm control system 81 to plan a path for the unloading arm.

[0118] ​In the embodiment of the present application, the control system 8 further comprises a flange interface position prediction system 83 for predicting the trend of the displacement or deformation of the flange interface position based on the ship body movement angle information collected by the image recognition module 6, simulating the ship body sway caused by sea waves.

[0119] The flange interface position prediction system 83 is installed on the upper computer.

[0120] The input end of the flange interface position prediction system 83 is electrically connected with the motion control system 82, and the output end is electrically connected with the unloading arm control system 81, and the connection mode can adopt wireless connection mode or wired connection mode.

[0121] Specifically, the intelligent identification model is also equipped with a roboflow tool, which calibrates the target flange interface position in the data set collected by the image recognition module 6 through the roboflow tool, and transmits the calibrated data set to the flange interface position prediction system 83. The flange interface position prediction system 83 collects the change position data of the target flange interface in the continuous frames, selects the LSTM deep learning method, trains the position prediction model according to the change position data, inputs the latest flange interface image into the position prediction model, and outputs the future position of the target flange interface relative to the unloading arm.

[0122] Further, the calibrated data set is input to the flange interface position prediction system 83 through the motion control system 82, and the prediction result is sent to the unloading arm control system 81.

[0123] In the embodiment, by analyzing the position change of the target flange interface in the image data set, the influence law of the current wind and wave on the ship body is accurately obtained, so as to realize the prediction of the subsequent ship body position, help the unloading arm intelligent docking device to adjust the motion posture of the unloading arm in advance, improve the operation stability, optimize the docking path, reduce the wind and wave interference, further improve the docking efficiency, enhance the anti-interference ability of the unloading arm docking device, reduce the influence of wind and wave on the unloading arm docking precision, and help to complete the docking task of the unloading arm more safely and efficiently.

[0124] In the embodiment of the present application, the driving module 9 comprises an inner arm driving module 91, an outer arm driving module 92 and a horizontal driving module 93.

[0125] The inner arm driving module 91 is used to drive the inner arm 3 joint of the unloading arm to move vertically based on the control instruction.

[0126] The outer arm driving module 92 is used to drive the outer arm 4 joint of the unloading arm to move vertically based on the control instruction.

[0127] The horizontal driving module 93 is used to drive the inner arm 3 joint of the unloading arm to move horizontally based on the control instruction.

[0128] In the embodiment, the inner arm driving module 91, the outer arm driving module 92 and the horizontal driving module 93 are arranged, which can significantly improve the flexibility and adaptability of the movement of the unloading arm, so that it can accurately cope with complex three-dimensional space operation requirements, and through independent adjustment of the horizontal and vertical directions, more accurate unloading arm docking is realized. At the same time, the design of multiple drivers simplifies the control logic, reduces the control complexity, improves the overall operation efficiency and reliability of the unloading arm, and ensures the efficiency, accuracy and safety of the docking operation.

[0129] The second aspect of the application provides an intelligent docking method of an unloading arm, and the method flow is as shown in Figure 3 The intelligent docking method of the unloading arm comprises the following steps:

[0130] S1: Collecting a first flange interface position image, and obtaining first position information of a target flange interface according to the first flange interface position image.

[0131] Wherein, multiple first flange interface position images are collected to form a first flange interface position image dataset, and the intelligent recognition model analyzes the first position information according to the first flange interface position image dataset.

[0132] S2: Controlling the driving module 9 to drive the unloading arm to complete the preliminary docking according to the first position information.

[0133] S3: Collecting a second flange interface position image, and obtaining second position information of the target flange interface according to the second flange interface position image.

[0134] Wherein, multiple second flange interface position images are collected to form a second flange interface position image dataset, and the intelligent recognition model analyzes the second position information according to the second flange interface position image dataset.

[0135] S4: Collecting distance information of the target flange interface from the unloading arm.

[0136] S5: Controlling the driving module 9 to drive the unloading arm to complete the secondary docking according to the second position information and the distance information.

[0137] Specifically, the intelligent docking method of the unloading arm is mainly implemented based on the unloading arm control system 81, after collecting the position information and angle information of the target flange interface, the current angle of each joint of the unloading arm is also collected through the rope sensor 71 to determine the current posture of the unloading arm.

[0138] The unloading arm control system 81 generates control instructions according to the position information, distance information and current angle of each joint of the unloading arm, and sends the control instructions to the motion control system 82, and the motion control system 82 controls the driving module 9 to drive the unloading arm to complete the preliminary docking and the secondary docking based on the control instructions.

[0139] The control instructions include a preliminary docking control instruction and a secondary docking control instruction, which are respectively used for the unloading arm to perform preliminary docking and secondary docking.

[0140] In the primary docking process, the monocular camera 61 is used to select the target flange interface and obtain its first position information, which is fed back to the unloading arm control system 81. The unloading arm control system 81 generates a preliminary docking control instruction for controlling the movement of the unloading arm through path planning, and the preliminary docking is completed when the three-dimensional joint 51 of the unloading arm is 1 m away from the target flange interface.

[0141] After the unloading arm completes the preliminary docking, the monocular camera 61 located on the column 2 is blocked by the three-dimensional joint 51 and cannot observe the position of the flange. Therefore, the secondary docking needs to rely on the binocular depth camera and the ultrasonic sensor to complete. In order to reduce the precision error caused by motion inertia, the secondary docking adopts a slow, small distance and multiple movement mode. By adding a motion delay function in the YOLO algorithm, the influence of shaking on the docking work is reduced. When the three-dimensional joint 51 is 40 cm away from the target flange interface, the three-dimensional joint claw 52 and the guide plate 53 are automatically opened. When the three-dimensional joint 51 is 15 cm away from the target flange interface, the binocular camera 62 cannot capture the target flange interface. At this time, the unloading arm pose is adjusted by the ultrasonic sensor 73 to ensure the accuracy of the docking and achieve precise docking. After the manifold 54 completely enters the target flange interface, the three-dimensional joint claw 52 closes and locks the target flange interface, completing the secondary docking.

[0142] The motion delay function added in the YOLO algorithm makes the unloading arm stay at the current position for a certain amount of time after each movement is completed, thereby reducing the shaking amplitude in the movement of the unloading arm, ensuring that the constituent structure of the unloading arm is in a stable state, avoiding the influence of the structure shaking caused by the continuous output of the hydraulic system on the docking work.

[0143] In this embodiment, through the two processes of preliminary docking and secondary docking, the running efficiency, safety and precision of the unloading arm docking are comprehensively improved. The preliminary docking is responsible for quickly guiding the unloading arm to the target area, greatly improving the initial docking speed and reducing the running difficulty. The secondary docking completes the final high-precision connection on the basis of the preliminary docking, ensuring the accuracy and stability of the docking. The step-by-step operation enhances the adaptability of the unloading arm system in complex working conditions, optimizes the resource utilization, reduces the requirements for equipment and algorithms, and the unloading arm intelligent docking process does not require manual intervention. The operator can complete the docking through computer control, significantly reducing the labor intensity and the complexity and workload of manual operation.

[0144] In the embodiments of the present application, the driving module is controlled to drive the unloading arm to complete the preliminary docking according to the first position information, and the method further comprises:

[0145] Real-time acquisition of ship motion angle information is used to real-time predict the position of the target flange interface, and based on the prediction result, the first position information and the distance information, a preliminary docking control instruction is generated, and the driving module 9 is controlled to drive the unloading arm to complete the preliminary docking according to the preliminary docking control instruction.

[0146] The driving module 9 is controlled to drive the unloading arm to complete the secondary docking according to the second position information and the distance information, and the method further comprises:

[0147] Real-time acquisition of ship motion angle information is used to real-time predict the position of the target flange interface, and based on the prediction result, the second position information and the distance information, a secondary docking control instruction is generated, and the driving module 9 is controlled to drive the unloading arm to complete the secondary docking according to the secondary docking control instruction.

[0148] In the preliminary docking and the secondary docking process, the ship motion angle information is obtained by the intelligent recognition model constructed by the image recognition module 6, based on the first flange interface position image data set and the second flange interface position image data set, and the ship position prediction is performed by the flange interface position prediction system 83.

[0149] In the embodiments, the ship motion state is accurately obtained by real-time acquisition of the ship motion angle information, the unloading arm is re-planned based on the ship motion state, the current working condition is accurately fitted, the safety and accuracy of the unloading arm docking are greatly improved, and the collision and connection point deviation are effectively prevented. At the same time, the unloading arm motion path is dynamically optimized, the repeated adjustment caused by inaccurate path is reduced, and the docking efficiency of the unloading arm is improved.

[0150] The above has described the embodiments of the present application, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles, practical applications or improvements to the technology in the market of the embodiments, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. An intelligent docking device for unloading arms, characterized in that: The device includes an image recognition module, a sensor module, a control system, and a drive module; The image recognition module, the sensor module, and the drive module are mounted on the unloading arm; The image recognition module is used to collect images of the target flange interface and information on the ship's motion angle in real time. The image recognition module includes a monocular camera and a binocular camera; The monocular camera is mounted on the column of the unloading arm to acquire the image of the first flange interface position, and selects the target flange interface to be docked by the unloading arm through an intelligent recognition model to obtain the first position information of the target flange interface. The binocular camera is installed on the three-dimensional joint of the unloading arm to acquire images of the second flange interface position. The intelligent recognition model is used to accurately determine the position of the target flange interface that the unloading arm needs to dock with, and obtain the second position information of the target flange interface. The image recognition module is equipped with an intelligent recognition model, which is used to analyze the relative position of the unloading arm and the target flange interface based on the target flange interface image collected by the image recognition module. After acquiring the target flange interface image, the image recognition module extracts the required images frame by frame and creates a dataset. After labeling the dataset, the intelligent recognition model is trained using the YOLO algorithm. The control system completes the initial docking of the target flange interface based on the first position information, and completes the secondary docking of the target flange interface based on the second position information; Specifically, a monocular camera is used to capture images of the target flange interface to obtain a first flange interface image dataset. The first position information of the target flange interface is obtained by analyzing the first flange interface image dataset based on an intelligent recognition model. Open the binocular camera to capture images of the target flange interface, obtain a second flange interface image dataset, and analyze the second flange interface image dataset based on the intelligent recognition model to obtain the second position information of the target flange interface. The sensor module is used to collect the motion angle information of the unloading arm in real time; The sensor module includes a pull rope sensor, a tilt sensor, and an ultrasonic sensor. The pull rope sensor is installed on the column shaft box of the unloading arm and at the connection of the three-dimensional joint, and is used to collect the horizontal rotation angle of the unloading arm and the horizontal rotation angle of the three-dimensional joint. The tilt sensor is installed on the inner and outer arms of the unloading arm to collect the joint angles of the inner and outer arms in real time. The ultrasonic sensor is installed on the three-dimensional joint of the unloading arm and is used to collect the distance information between the target flange interface and the unloading arm. The control system is used to monitor the movement of the unloading arm in real time and issue control commands based on the data collected by the image recognition module and the sensor module; The drive module is installed at the joint of the unloading arm. The drive module is used to receive control commands and drive the unloading arm to move to the target flange interface based on the control commands.

2. The intelligent docking device for unloading arms according to claim 1, characterized in that: The control system is also used for: A reference coordinate system for the unloading arm is established based on the base of the unloading arm; A DH coordinate system is established based on the unloading arm reference coordinate system to represent the movement direction of each joint of the unloading arm; A coordinate system for the three-dimensional joint of the unloading arm is established based on the reference coordinate system of the unloading arm, which is used to represent the position of the three-dimensional joint relative to the base; Based on the DH coordinate system and the coordinate system of the three-dimensional joint, the inverse kinematic equations are constructed, and the motion angles of each joint of the unloading arm when it runs to the target flange interface position are obtained by combining the mathematical twin model.

3. The intelligent docking device for unloading arms according to claim 1, characterized in that, The control system includes a discharge arm control system and a motion control system; The unloading arm control system is installed on the host computer and is used to monitor the operating status of the unloading arm and generate control commands to control the operation of the unloading arm. The motion control system is installed on the lower-level computer and is used to receive data collected by the image recognition module and the sensor module, and to control the drive module according to the control instructions.

4. The intelligent docking device for unloading arms according to claim 3, characterized in that, The control system also includes a flange interface position prediction system, which is used to simulate the hull sway caused by the influence of sea waves based on the hull motion angle information collected by the image recognition module, and predict the offset or deformation trend of the flange interface position.

5. The intelligent docking device for unloading arms according to claim 1, characterized in that, The drive module includes an inner arm drive module, an outer arm drive module, and a horizontal drive module. The inner arm drive module is used to drive the inner arm joint of the unloading arm to move vertically based on the control command; The outer arm drive module is used to drive the outer arm joint of the unloading arm to move vertically based on the control command; The horizontal drive module is used to drive the inner arm joint of the unloading arm to move horizontally based on the control command.

6. A method for intelligent docking of a discharge arm, applied to the intelligent docking device for a discharge arm as described in any one of claims 1-5, characterized in that, The intelligent docking method for the unloading arm includes: Acquire an image of the first flange interface position, and obtain the first position information of the target flange interface based on the first flange interface position image; Based on the first position information, the drive module is controlled to drive the unloading arm to complete the initial docking; Acquire an image of the second flange interface position, and obtain the second position information of the target flange interface based on the image of the second flange interface position; Collect the distance information between the target flange interface and the unloading arm; Based on the second position information and the distance information, the drive module is controlled to drive the unloading arm to complete the secondary docking.

7. The intelligent docking method for unloading arms according to claim 6, characterized in that, Based on the first position information, the drive module is controlled to drive the unloading arm to complete the initial docking, and the following is also included: The ship's motion angle information is collected in real time to predict the position of the target flange interface. Based on the prediction result, the first position information and the distance information, a preliminary docking control command is generated. The drive module is controlled to drive the unloading arm to complete the preliminary docking according to the preliminary docking control command. Based on the second position information and the distance information, the drive module is controlled to drive the unloading arm to complete the secondary docking, and the method further includes: The ship's motion angle information is collected in real time to predict the target flange interface position. Based on the prediction result, the second position information, and the distance information, a secondary docking control command is generated. The drive module is controlled to drive the unloading arm to complete the secondary docking according to the secondary docking control command.

Images