Compensation device and compensation method for balance of pose of end object hoisted by ship-mounted crane

Abstract

The present application relates to the technical field of ship-mounted cranes, and provides a compensation device and a compensation method for object pose balance at a hoisting end of a ship-mounted crane. The compensation device for object pose balance at the hoisting end of the ship-mounted crane comprises a crane, a detection assembly and a controller. The crane is arranged on a deck platform of a ship, and the hoisting end of the crane hoists an object through a plurality of hoisting ropes. The detection assembly is arranged on the deck platform of the ship, and is used to acquire the pose of the ship and image information of the object in real time. The controller is used to control the rotation of each joint of the crane and adjust the length of each hoisting rope, so that the object is in a balanced state. The compensation device for object pose balance at the hoisting end of the ship-mounted crane can automatically adjust the position of the object and the length of each hoisting rope when the ship rolls, pitches and heaves, so that the position and the attitude of the object can be in a balanced state in real time, and the stability of the crane when hoisting the object on the ship is improved.

Description

Technical Field

[0001] This invention relates to the field of shipborne crane technology, and in particular to a compensation device and method for balancing the position and posture of an object at the lifting end of a shipborne crane. Background Technology

[0002] In harsh marine environments, ships experience six degrees of freedom of motion: sway, pitch, heave, roll, pitch, and bow. The objects lifted by shipboard cranes will also experience violent movements. Failure to adjust the attitude of the lifted objects in a timely manner can easily lead to serious accidents. While ships can maintain balance by adjusting their attitude when experiencing sway, pitch, and bow, the position of the objects lifted by cranes cannot be guaranteed to remain balanced when a ship experiences roll, pitch, and heave in rough seas.

[0003] There is currently a real-time compensation control method for the lifting end position of a shipborne crane based on inverse kinematics. It uses an inertial navigation sensor to detect the real-time attitude signal of the ship's motion, establishes a global coordinate system for the entire system platform, calculates the position change of the lifting end of the shipborne robotic arm based on the attitude signal and the homogeneous transformation of each coordinate system, and solves the rotation angle of the robotic arm joint through an inverse kinematics algorithm.

[0004] Another active wave compensation method is based on binocular vision. This method equips the crane with eight servo motors with wire ropes and two cameras. The two cameras detect the pose of the object at the lifting end, and calculate the direction and rotation angle of the servo motors based on the detected deviation in the object's pose. The servo motors then adjust the object's pose according to the direction and rotation angle.

[0005] Currently, most cranes use a single lifting rope to lift heavy objects, which makes it difficult to guarantee the balance of the object's posture under the influence of waves. Two main methods exist: compensation based on information from ship attitude sensors and compensation based on machine vision.

[0006] Compensation is based on information from ship attitude sensors. The rotation angle of the crane joint is calculated by monitoring the ship's motion state. Although the positional offset of the object can be calculated, the position and posture of the object at the lifting end cannot be guaranteed to be balanced when controlling the rotation of the crane joint. At the same time, the object will have an unbalanced posture due to the influence of sea conditions, and the posture of the object cannot be adjusted.

[0007] Machine vision compensation methods are effective when the swing amplitude of the object at the lifting end is small. However, when the wind and waves are large, the swing amplitude and frequency of the object at the lifting end increase, and machine vision often cannot quickly identify the pose of the object at the lifting end. If the object cannot be identified, it will seriously affect the compensation effect on the object at the lifting end. Therefore, it is essential to provide a new compensation device and method for multi-point lifting of objects at the lifting end of a crane. Summary of the Invention

[0008] This invention provides a compensation device and method for the positional balance of an object at the lifting end of a shipborne crane, in order to solve the defect in the prior art where the positional balance of an object at the lifting end of a crane is difficult to adjust.

[0009] This invention provides a compensation device for the positional balance of an object at the lifting end of a shipborne crane, comprising: a crane, mounted on a ship's deck platform, wherein the lifting end of the crane lifts an object via multiple lifting ropes; a detection component, mounted on the ship's deck platform, for acquiring real-time image information of the ship's position and the object; and a controller, for controlling the rotation of each joint of the crane according to the ship's current position to adjust the offset position of the object, the controller also for calculating the distance of each lifting point relative to the same horizontal plane based on the image information of the object, and adjusting the length of each lifting rope according to the distance to keep the object in a balanced state.

[0010] According to the present invention, a compensation device for the position and posture balance of an object at the lifting end of a shipborne crane is provided. The detection component includes: an attitude sensor disposed on the ship's deck platform, the attitude sensor being used to acquire the ship's position and posture in real time; and an image acquisition element disposed on the ship's deck platform, the image acquisition element being used to acquire image information of the object in real time.

[0011] The present invention also provides a compensation method using the compensation device for the position and posture balance of an object at the lifting end of a shipborne crane as described above, comprising: acquiring the position and posture of the ship in real time, controlling the rotation of each joint of the crane based on the position and posture to adjust the offset position of the object; acquiring image information of the object in real time, predicting the first spatial coordinates of each lifting point when the object is in a balanced state again based on the image information; acquiring the current image information of the object, obtaining the second spatial coordinates of each lifting point based on the current image information; and adjusting the length of each lifting rope based on the difference between the second spatial coordinates and the first spatial coordinates.

[0012] According to the present invention, a method for compensating for the positional balance of an object at the lifting end of a shipborne crane is provided. The step of acquiring the positional balance of the ship in real time and controlling the rotation of each joint of the crane based on the positional balance to adjust the offset position of the object further includes: calculating the compensation angle of each joint of the crane based on the positional balance of the ship by inverse kinematics; and controlling the rotation of each joint of the crane based on the compensation angle.

[0013] According to the present invention, a method for compensating for the pose balance of an object at the lifting end of a shipborne crane, the step of acquiring image information of the object in real time and predicting the first spatial coordinates of multiple lifting points when the object is in a balanced state again based on the image information further includes: acquiring red, green and blue images of the object image information, and predicting the first spatial coordinates of each lifting point when the object is in a balanced state again after processing by a network model.

[0014] According to the present invention, a method for compensating for the positional balance of an object at the lifting end of a ship-mounted crane, wherein the step of adjusting the length of each lifting rope based on the difference between the second spatial coordinate and the first spatial coordinate further includes: obtaining the coordinates of the first center point of the plane where each lifting point is located when the object is in a balanced state before lifting; obtaining the coordinates of the second center point of the plane where each lifting point is located when the object is in a balanced state again; and calculating the actual spatial coordinates of each lifting point when the object is in a balanced state based on the relationship that the distance between each lifting point and the second center point is equal to the distance between each lifting point and the first center point.

[0015] According to the present invention, a method for compensating for the positional balance of an object at the lifting end of a shipborne crane includes a step of adjusting the length of each lifting rope based on the difference between the second spatial coordinate and the first spatial coordinate. The step further includes: obtaining the first length of each lifting rope when the object is in a balanced state before lifting; calculating the second length of each lifting rope based on the actual spatial coordinate; and calculating the adjustment amount of each lifting rope based on the difference between the second length and the first length.

[0016] According to the present invention, a method for compensating for the pose balance of an object at the lifting end of a ship-mounted crane, the step of obtaining the coordinates of the first center point of the plane containing multiple lifting points when the object is in a balanced state before lifting further includes: obtaining a red-green-blue image of the object when it is in a balanced state before lifting, and obtaining the third spatial coordinates of each lifting point of the object after processing by a network model; and calculating the coordinates of the first center point based on the third spatial coordinates.

[0017] According to the present invention, a method for compensating for the pose balance of an object at the lifting end of a ship-mounted crane is provided. The step of acquiring image information of the object in a balanced state before lifting and obtaining the third spatial coordinates of each lifting point of the object based on the image information of the object in a balanced state before lifting further includes: acquiring red-green-blue images of the image information of the object in a balanced state before lifting, and obtaining the third spatial coordinates of each lifting point of the object after processing by a network model.

[0018] According to the present invention, a method for compensating for the positional balance of an object at the lifting end of a shipborne crane, the step of obtaining the coordinates of the second center point of the plane containing the multiple lifting points when the object is in a balanced state further includes: calculating the coordinates of the second center point based on the first spatial coordinates.

[0019] The present invention provides a compensation device for the position and posture balance of objects at the lifting end of a shipborne crane. By setting up detection components and a controller, and using multiple lifting ropes to lift objects, it can automatically adjust the position of the object and the length of each lifting rope by adjusting the rotation angle of each joint of the crane and the length of each lifting rope when the ship is rolling, pitching, or heaving. This ensures that the position and posture of the object are in a balanced state in real time, improves the stability of the crane when lifting objects on the ship, effectively suppresses the swaying of the object, reduces the risk of damage to the object under adverse sea conditions, and is conducive to production and widespread application. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 This is a structural schematic diagram of the compensation device for balancing the position and posture of an object at the lifting end of a shipborne crane provided by the present invention.

[0022] Figure 2 This is a schematic diagram of the operation of the compensation device for the position and posture balance of the object at the lifting end of the ship-borne crane provided by the present invention;

[0023] Figure 3 A flowchart of the compensation method for the positional balance of an object at the lifting end of a shipborne crane provided by the present invention;

[0024] Figure 4 A coordinate system diagram of the crane;

[0025] Figure 5 A diagram showing the rotation angle of the first joint of a crane;

[0026] Figure 6 A diagram showing the rotation angles of the second and third joints of a crane;

[0027] Figure label:

[0028] 10: Crane; 11: First joint; 12: Second joint; 13: Third joint; 14: Winch; 21: Attitude sensor; 22: Image acquisition element; 30: Lifting rope; 100: Ship deck platform. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0030] The terms "first" and "second" in the specification and claims of this invention may explicitly or implicitly include one or more of those features. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0031] The following is combined with Figures 1-6 This invention describes a compensation device and method for balancing the position and posture of an object at the lifting end of a shipborne crane.

[0032] like Figure 1 and Figure 2 As shown, in an embodiment of the present invention, the compensation device for the positional balance of an object at the lifting end of a shipborne crane includes: a crane 10, a detection component, and a controller. The crane 10 is mounted on a ship's deck platform 100, and the lifting end of the crane 10 lifts an object via multiple lifting ropes 30. The detection component is mounted on the ship's deck platform 100 and is used to acquire real-time image information of the ship's position and the object. The controller is used to control the rotation of each joint of the crane 10 according to the ship's current position to adjust the offset position of the object; the controller is also used to calculate the distance of each lifting point relative to the same horizontal plane based on the image information of the object, and adjust the length of each lifting rope 30 according to this distance to keep the object in a balanced state.

[0033] Specifically, when an object is lifted, the object's position changes when the ship rolls, pitches, or heaves, causing the object to lose its balance. Based on this, the object position balance compensation device at the lifting end of the shipborne crane provided in this embodiment of the invention can adjust the object's position in real time when the ship rolls, pitches, or heaves, so that the object is restored to a balanced state.

[0034] In this embodiment, the crane 10 includes three boom sections. The first end of the first boom section is rotatably connected to the chassis of the crane 10 via a first joint 11. The second end of the first boom section is rotatably connected to the first end of the second boom section via a second joint 12. The second end of the second boom section is rotatably connected to the first end of the third boom section via a third joint 13. The second end of the third boom section is connected to a winch 14. Four lifting ropes 30 are connected to the bottom surface of the winch 14 in a circumferential direction. The lifting ropes 30 are used to lift objects.

[0035] When the ship rolls, pitches, or heaves, the lifting ropes 30 swing accordingly, causing the object to deviate from its original position. At this time, the detection component acquires the ship's current orientation. Based on this orientation, the controller calculates the compensation angles of each joint of the crane 10 using inverse kinematics. By adjusting the rotation angles of each joint, the object's position can be changed, aligning it with the winch 14. However, the lifting points may not be on the same plane at this point. The detection component acquires real-time image information of the object. Based on this image information, the controller predicts the first spatial coordinates of each lifting point when the object returns to equilibrium. Simultaneously, the controller obtains the second spatial coordinates of each lifting point based on the current image information. Comparing the second spatial coordinates with the first spatial coordinates yields the adjustment amount of each lifting rope 30. The controller adjusts the length of each lifting rope 30 according to this adjustment amount, thereby restoring the object to equilibrium.

[0036] Furthermore, in this embodiment, after the controller controls the joints of the crane 10 to rotate, the position of the object can be adjusted. At this time, the x-axis and y-axis coordinates in the second spatial coordinates of each lifting point are the same as the x-axis and y-axis coordinates in the first spatial coordinates. The difference between the z-axis coordinate in the second spatial coordinates and the z-axis coordinate in the first spatial coordinates is the adjustment amount of each lifting rope 30.

[0037] The compensating device for the position and posture balance of objects at the lifting end of a shipborne crane provided in this embodiment of the invention, by setting up detection components and a controller, and using multiple lifting ropes to lift objects, can automatically adjust the position of the object and the length of each lifting rope by adjusting the rotation angle of each joint of the crane and the length of each lifting rope when the ship is rolling, pitching, or heaving. This allows the position and posture of the object to be kept in a balanced state in real time, improving the stability of the object when the crane is lifting objects on the ship, effectively suppressing the swaying of the object, reducing the risk of damage to the object under the influence of severe sea conditions, and facilitating production and widespread application.

[0038] like Figure 1As shown, in an embodiment of the present invention, the detection component includes an attitude sensor 21 and an image acquisition element 22. Both the attitude sensor 21 and the image acquisition element 22 are mounted on the ship's deck platform 100. The attitude sensor 21 is used to acquire the ship's pose in real time, and the image acquisition element 22 is used to acquire image information of the object in real time. Optionally, in an embodiment of the present invention, the image acquisition element 22 can be a high-definition camera.

[0039] like Figure 3 As shown in the figure, this embodiment of the invention also provides a method for compensating for the positional balance of an object at the lifting end of a shipborne crane, specifically including the following steps:

[0040] Step 01: Acquire the ship's pose in real time, and control the rotation of each joint of the crane 10 based on the ship's pose to adjust the offset position of the object; Step 02: Acquire the image information of the object in real time, and predict the first spatial coordinates of each lifting point when the object is in a balanced state again based on the image information; Step 03: Acquire the current image information of the object, and obtain the second spatial coordinates of each lifting point based on the current image information; Step 04: Adjust the length of each lifting rope based on the difference between the second spatial coordinates and the first spatial coordinates.

[0041] Specifically, when the ship rolls, pitches, and heaves, the hoisting ropes 30 swing accordingly, causing the object to deviate from its original position. The attitude sensor 21 acquires the ship's current attitude. Based on this attitude, the controller calculates the compensation angles of each joint of the crane 10 using inverse kinematics. By adjusting the rotation angles of each joint, the object's position can be changed, positioning it relative to the winch 14. The image acquisition element 22 acquires real-time image information of the object. Based on this image information, the controller predicts the first spatial coordinates of each lifting point when the object is back in equilibrium. Simultaneously, based on the current image information, the controller obtains the second spatial coordinates of each lifting point. Comparing the second and first spatial coordinates yields the adjustment amount of each hoisting rope 30. The controller adjusts the length of each hoisting rope 30 according to this adjustment amount, thus restoring the object to equilibrium.

[0042] Furthermore, in this embodiment, after the controller controls the joints of the crane 10 to rotate, the position of the object can be adjusted. At this time, the x-axis and y-axis coordinates in the second spatial coordinates of each lifting point are the same as the x-axis and y-axis coordinates in the first spatial coordinates. The difference between the z-axis coordinate in the second spatial coordinates and the z-axis coordinate in the first spatial coordinates is the adjustment amount of each lifting rope 30.

[0043] The compensation method for the position and posture balance of an object at the lifting end of a shipborne crane provided in this invention acquires the ship's position and posture in real time, and obtains the compensation angle of each joint of the crane based on the ship's position and posture. The rotation of each joint is controlled according to the compensation angle, thereby adjusting the position of the object. By acquiring the image information of the object in real time, the first spatial coordinates when the object is in a balanced state and the second spatial coordinates after the object's position adjustment can be predicted. By comparing the second spatial coordinates with the first spatial coordinates, the adjustment amount of each lifting rope is obtained, so that the position and posture of the object can be kept in a balanced state in real time. This improves the stability of the object being lifted by the crane on the ship, effectively suppresses the swaying of the object, and reduces the risk of damage to the object under the influence of severe sea conditions.

[0044] Furthermore, in an embodiment of the present invention, the step of acquiring the ship's pose in real time and controlling the rotation of each joint of the crane 10 based on the pose to adjust the offset position of the object further includes: calculating the compensation angle of each joint of the crane 10 based on the ship's pose by inverse kinematics formula; and controlling the rotation of each joint of the crane 10 based on the compensation angle.

[0045] Specifically, such as Figure 4 As shown, a coordinate system is established with the three joints of the crane 10 and the second end of the third boom as the center. In this embodiment, the default included angle between the first joint 11 and the second joint 12 of the crane 10 is 180°. The z-axis of the spatial coordinate system is the rotation axis of the joint, the x-axis is a straight line perpendicular to the rotation axis of this joint and the rotation axis of the next joint, and the y-axis is perpendicular to the x-axis and the z-axis. The four coordinate systems {0}, {1}, {2}, and {3} are established. Figure 4 The coordinate system shown allows us to obtain the DH parameter table for crane 10. The DH parameters (Denavit–Hartenberg parameters) are a mathematical model and coordinate system for determining the positional and angular relationships between two pairs of boom segments, expressed using four parameters. See Table 1 for details.

[0046] Table 1 Crane DH Table

[0047] i θ d a α 1 0° L1 0 90° 2 90° 0 L2 0° 3 -90° 0 L3 0°

[0048] Where θ represents the angle between the Z axes of adjacent coordinate systems, d represents the distance between the two X axes of adjacent coordinate systems, d represents the distance between the two Z axes of adjacent coordinate systems, and α represents the angle between the Z axes of adjacent coordinate systems.

[0049] set up Let be the homogeneous coordinate transformation matrix for translation and rotation from coordinate system a to b. Let (x, y, z) represent the set of translation vectors of b relative to joint a. This represents the rotational transformation of the crane's end-effector posture at joint angle θi (i = 1, 2, 3, 4):

[0050]

[0051] The parameters can then be obtained from the crane DH parameter table. Then we have:

[0052]

[0053] by Figure 4 The coordinate system in which the first joint 11 of the crane 10 is located is the base coordinate system. This refers to the pose of the crane 10 lifting end relative to the base coordinate system. The rotation axes and directions of the simulated six-degree-of-freedom platform are defined, with the center of the ship's deck platform 100 plane as the center, the direction perpendicular to the platform plane as the z-axis, the roll direction as the x-axis, and the pitch direction as the z-axis. P is used as the coordinate system. t R represents the heave z-vector of the simulated platform; α represents the set of roll φ and pitch Θ vectors; t (α) represents the rotational transformation of the ship simulation platform at time t when the disturbance attitudes are φ and Θ. Let c and s denote the abbreviations for cos and sin, respectively. Then, the homogeneous transformation matrix of the ship's base coordinate system {bt-1} at time t-1 relative to the ship's base coordinate system {bt} at time t is:

[0054]

[0055] Multiplying (3) by (2) on the left yields the pose of the crane 10 lifting end object relative to the base coordinates under the current ship attitude. Let θi represent the rotation angle of the i-th joint, and (x,y,z) represent the lifting end coordinates. The offset difference of the z-axis in the spatial coordinate system is compensated by the lifting rope 30. L is the length of the lifting rope 30, and z t z0 represents the z-axis coordinate of the object at the current time t and the z-axis coordinate of the object at the initial position. We can obtain...

[0056] L = z t -z0 (4)

[0057] By calculating the inverse kinematics based on the offsets along the x and y axes, the rotation angles θ1, θ2, and θ3 of each joint after compensation can be obtained. The inverse kinematics of the robot is to calculate the rotation angles of each joint of the crane 10 based on the pose of the lifting end of the crane 10 relative to the base coordinates. Let θi represent the rotation angle of the i-th joint, and (x, y, z) represent the coordinates of the lifting end.

[0058] Solve for θ1:

[0059] Based on the projection of x and y onto the xoy plane, and the x and y values ​​of the lifting end coordinates of crane 10, as follows: Figure 5As shown, the rotation angle of the first joint 11 can be obtained:

[0060] θ1=atan2(y,x) (5)

[0061] Solve for θ3:

[0062] Solving for θ3 can be done indirectly by finding its supplementary angle ψ. According to the Law of Cosines, to obtain ψ, we need information about the triangle formed by ψ and the second and third arm segments from the second joint 12 to the third joint 13, such as... Figure 6 As shown.

[0063] From DH table and Figure 1 We know that the lengths of the second and third arm segments are a2 and a3, respectively. The length of the hypotenuse k is:

[0064]

[0065] According to (6):

[0066]

[0067] Since ψ and θ3 are complementary, then:

[0068]

[0069] Solve for θ2:

[0070] set up Let ω be the angle formed by the line connecting the lifting end of crane 10 to the origin of the base coordinate system and the x0y plane, and let ω be the angle formed by the line connecting the second boom section and the lifting end of crane to the origin of the base coordinate system. Then we have:

[0071]

[0072] ω=atan2(L3*sinθ3,L2+L3*cosθ3) (10)

[0073] From (9) and (10) we can know:

[0074]

[0075] In summary, the rotation angles θ1, θ2, and θ3 of each joint are obtained.

[0076] Furthermore, in an embodiment of the present invention, the step of acquiring image information of the object in real time and predicting the first spatial coordinates of multiple lifting points when the object is in a balanced state again based on the image information further includes: acquiring red, green and blue images of the object image information, and predicting the first spatial coordinates of each lifting point when the object is in a balanced state again after processing by a network model.

[0077] Specifically, the red-green-blue image obtained by image acquisition element 22, i.e., the image with color (RGB color mode is an industry color standard, which obtains various colors by changing the three color channels of red (R), green (G), and blue (B) and superimposing them), is processed by the network model to obtain the 3D pose of the object. The Pix2Pose model can predict the pose of objects without texture, and convolutional neural networks (CNNs) can also estimate the pose of objects without texture. However, CNNs have lower accuracy when predicting symmetrical objects and occluded objects. The Pix2Pose model is more robust and can stably regress the pixel-level 3D coordinates of the object from the red-green-blue image. In this embodiment, the Pix2Pose model is selected. The specific process of real-time prediction of the 3D coordinates of the four lifting points of the object at the lifting end and compensation of the object's pose by the lifting rope 30 is as follows:

[0078] First, the 3D coordinates of the object's lifting point are predicted. Each frame of the image captured by the camera is inspected. The lifting ropes used for each lifting operation remain the same. The red, green, and blue values ​​of the pixels in the area occupied by the ropes differ significantly from other areas in the image. Faster R-CNN is an image detection algorithm model. First, the video of the object in the real-world scene is used to extract several image frames. The area occupied by the lifting rope 30 in the images is manually labeled and compiled into a dataset, which is then trained by Faster R-CNN. The trained algorithm model can accurately detect the area occupied by the lifting rope 30 in the image. Using the trained Faster R-CNN detection algorithm, the areas occupied by the four lifting ropes 30 are cropped based on pixel information and used as input images for the Pix2Pose model. The Pix2Pose model is a publicly available network model that can render the input image and output a rendered color image, establishing a 2D-3D correspondence based on the color image information. In this embodiment, the input image is a cropped image of the lifting ropes 30. All the ropes 30 have a uniform color. After rendering using the Pix2Pose model, the portion of the rope at different spatial positions will display different colors. The image is composed of countless pixels, and the red, green, and blue values ​​of each pixel correspond to the x, y, and z coordinates in a 3D coordinate system. This allows us to return the 3D coordinates of the boundaries of the lifting ropes 30. The upper boundaries of the four lifting ropes 30 correspond to the coordinates of the point where they converge with the winch, and the lower boundaries correspond to the coordinates of the lifting points that contact the object. Based on this method, we can predict the first spatial coordinates of multiple lifting points when the object is back in a balanced state, and we can also obtain the second spatial coordinates of multiple lifting points after the object's position has been adjusted.

[0079] Furthermore, the step of adjusting the length of each lifting rope 30 based on the difference between the second spatial coordinates and the first spatial coordinates further includes: obtaining the coordinates of the first center point of the plane where each lifting point is located when the object is in a balanced state before lifting; obtaining the coordinates of the second center point of the plane where each lifting point is located when the object is in a balanced state again; and calculating the actual spatial coordinates of each lifting point when the object is in a balanced state based on the relationship that the distance between each lifting point and the second center point is equal to the distance between each lifting point and the first center point.

[0080] Specifically, the red, green and blue images of the object in a balanced state before it is lifted are obtained. After processing by the network model, the third spatial coordinates of each lifting point of the object are obtained. The coordinates of the first center point are calculated based on the third spatial coordinates. The coordinates of the second center point are calculated based on the first spatial coordinates.

[0081] The steps of adjusting the length of each suspension rope 30 based on the difference between the second spatial coordinates and the first spatial coordinates further include: obtaining the first length of each suspension rope 30 when the object is in a balanced state before being lifted; calculating the second length of each suspension rope 30 based on the actual spatial coordinates; and calculating the adjustment amount of each suspension rope 30 based on the difference between the second length and the first length.

[0082] Specifically, the third spatial coordinates of the four lifting points where the object is in a balanced posture before lifting are recorded, as well as the first length of the four lifting ropes (30 mm). i0 =(X i0 ,Y i0 Z i0 )T(i=1,2,3,4),A0=(X A0 ,Y A0 Z A0 T(i=1,2,3,4), l i (i = 1, 2, 3, 4). At any time t, the second length of each suspension rope 30 that keeps the object in equilibrium is l. it (i = 1, 2, 3, 4). The first spatial coordinates of the four lifting points are Pi, ... it =(X it ,Y it Z it T(i=1,2,3,4), A t =(X At ,Y At Z At Given )T(i=1,2,3,4), calculate the pose O0 of the first center point of the plane containing the lifting point, using the following formula:

[0083]

[0084] From formula (12), we can obtain the coordinates O0 of the first center point of the plane where the lifting point is located under the initial equilibrium posture and the center O1 of the second center point of the plane where the lifting point is located at time t. When the object is in equilibrium, the distance from each lifting point to the second center point is equal to the distance from each lifting point to the first center point under the initial position. Based on this principle, we can calculate the actual spatial coordinates of each lifting point when the object's posture is in equilibrium.

[0085] P it ′=P i0 -O0+O1 (13)

[0086] The second length l of each suspension rope 30 is calculated based on the actual spatial coordinates. it (i = 1, 2, 3, 4)

[0087]

[0088] The difference L between the length of each suspension rope 30 at time t when the object's posture remains in balance, calculated based on the following formula (15), and the length of the four suspension ropes 30 before lifting, is the length of the suspension rope 30 for raising and lowering.

[0089] L = l it -l i (15)

[0090] In an embodiment of the present invention, the compensation method for the positional balance of an object at the lifting end of a shipborne crane further includes the following steps: controlling the movement of each lifting rope 30 according to the extension and retraction length of each lifting rope 30 so that the object is restored to a balanced state.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A compensation method for the positional balance of an object at the lifting end of a shipborne crane, characterized in that, include: The ship's position and attitude are acquired in real time. Based on the ship's position and attitude, the compensation angles of each joint of the crane are calculated by inverse kinematics. Based on the compensation angles, the rotation of each joint of the crane is controlled to adjust the offset position of the object. The red, green and blue images of the object are obtained, and after being processed by the network model, the first spatial coordinates of each lifting point when the object is in a balanced state are predicted, and the actual spatial coordinates of each lifting point when the object's posture is balanced are obtained. Obtain the current image information of the object, and based on the current image information, obtain the second spatial coordinates of each lifting point. Adjust the length of each suspension rope based on the difference between the second spatial coordinate and the actual spatial coordinate; The method for obtaining the actual spatial coordinates is as follows: The red, green and blue images of the object are obtained when it is in a balanced state before it is lifted. After processing by the network model, the third spatial coordinates of each lifting point of the object are obtained. Based on the third spatial coordinates, the coordinates of the first center point are calculated, wherein the formula for calculating the coordinates of the first center point is as follows: Where O0 is the first center point, Let O0 be the coordinate of P. 10 P 20 P 30 P 40 There are 4 lifting points; The red, green and blue images of the object are obtained and then processed by a network model to predict the first spatial coordinates of each lifting point when the object is in a balanced state again. Based on the first spatial coordinates, the coordinates of the second center point are calculated in the same way as the coordinates of the first center point. Based on the relationship that the distance between the actual spatial coordinates of each lifting point and the second center point when the object is in equilibrium is equal to the distance between each lifting point and the first center point when the object is in equilibrium before lifting, the actual spatial coordinates of each lifting point when the object is in equilibrium are calculated.

2. A compensation device for implementing the compensation method for balancing the position of an object at the lifting end of a shipborne crane as described in claim 1, characterized in that, include: A crane, installed on a ship's deck platform, lifts objects using multiple lifting ropes at its lifting end; A detection component is installed on the ship's deck platform, and the detection component is used to acquire the ship's pose and image information of objects in real time; The controller is used to control the rotation of each joint of the crane according to the current position of the ship in order to adjust the offset position of the object. The controller is also used to calculate the distance of each lifting point relative to the same horizontal plane according to the image information of the object, and adjust the length of each lifting rope according to the distance to keep the object in a balanced state.

3. The compensation device according to claim 2, characterized in that, The detection component includes: An attitude sensor is installed on the ship's deck platform and is used to acquire the ship's position and attitude in real time. An image acquisition element is installed on the ship's deck platform, and the image acquisition element is used to acquire image information of objects in real time.

Images