High-precision measurement and automatic alignment method applicable to ship blocks with no margin

Abstract

The present invention is a high-precision measurement and automatic butt-joint method applicable to ship blocks with no margin. First, automatic calibration is performed on the measuring device, and the steps of measuring the visible key points on the block butt-joint surface and the coordinate positions of each trolley target ball are carried out. Coarse alignment is performed on the fixed block and the moving block to obtain the initial rotation matrix and the initial translation matrix of the key point set on the block butt-joint surface, and the accurate rotation matrix and the accurate translation matrix are obtained. The moving block butt-joint method is determined, and based on the coordinate positions of the trolley target balls, the position of the trolley relative to the moving block is determined. By combining the accurate rotation matrix and the accurate translation matrix, the trolley motion planned trajectory is obtained. The trolley motion trajectory is compensated to obtain the compensated path butt-joint trajectory. The trolley position data measured by two measuring devices are fused, and the trolley motion trajectory is fed back for control. A high-precision measurement and automatic butt-joint method applicable to ship blocks with no margin is disclosed. The method of the present invention realizes the automatic butt-joint of ship blocks through the automatic calibration of the measuring device, data calculation, automatic attitude adjustment, compensation prediction and control feedback.

Description

Technical Field

[0001] The present invention relates to the technology of automatic butt joint of ship blocks, and more specifically to a high-precision measurement and automatic butt joint method adapted to ship blocks.

Background Art

[0002] In the shipbuilding industry, the butt joint of ship blocks is directly related to the overall shipbuilding level. The quality of welding depends greatly on the accuracy of the butt joint of ship blocks. There are many factors affecting ship blocks. Among them, the three influencing factors of measurement error, attitude calculation error and attitude adjustment error are the most significant. Most patents reduce errors by proposing attitude measurement methods. For example, Chinese Patent CN200810229112.9 discloses a method for butt joint of two blocks during shipbuilding, and measures the coplanarity of the joint through a theodolite. However, it does not consider the mounting error of the device itself, does not consider the situation where important points on the block are shielded during measurement, and does not consider data conversion error, thus affecting the butt joint accuracy of ship blocks.

Summary of the Invention

Problems to be Solved by the Invention

[0003] Object of the Invention: The object of the present invention is to provide a high-precision measurement and automatic butt joint method adapted to ship blocks, which is an automatic butt joint method based on ship block measurement, data calculation, automatic attitude adjustment, compensation prediction and control feedback.

Means for Solving the Problems

[0004] Technical Solution: A high-precision measurement and automatic butt joint method adapted to the non-redundant blocks of a ship according to the present invention, Fix the fixed section and place it on a trolley with an adjustable moving section. Provide common reference points on the ground on both sides of the fixed section and the moving section. Provide corresponding key points on the butting surfaces of the fixed section and the moving section. Install a measuring device including a first laser tracker, a second laser tracker, and a laser distance meter, and calibrate the measuring device with the common reference points, calculate the position of the measuring device in the global coordinate system, measure the coordinates of each key point using the first laser tracker, measure the three-degree-of-freedom moving plane position coordinates of the trolley using the second laser tracker, and convert both the coordinates of each key point and the three-degree-of-freedom moving plane position coordinates of the trolley into the global coordinate system in step S1. Perform rough alignment on the fixed section and the moving section, and obtain the initial rotation matrix and the initial translation matrix of the key point set on the butting surface of the fixed section and the moving section. Step S2 is to obtain a more accurate rotation matrix and a more accurate translation matrix using the iterative closest point ICP algorithm. With each key point of the moving section as the center, draw a circle with a preset distance as the radius, calculate the weight of each key point of the moving section based on the number of pipes in each circle, select three key points with the maximum sum of weights and not collinear to form a plane, find the centroid of this plane, and based on the concept of random combinations, select three key points with the maximum sum of weights and not collinear other than the above set of combinations to form a plane, find the centroid of this plane, and sequentially find the centroids corresponding to the key points of the moving section in this way. Find the centroids corresponding to the key points on the fixed section in the same way, select three pairs of corresponding and non-collinear centroids on the fixed section and the moving section, establish the butting surface coordinate system of the moving section and the butting surface coordinate system of the fixed section respectively through the three-point alignment method, calculate the two coordinate systems, obtain the initial rotation matrix and the initial translation matrix, and realize the rough alignment between the moving section and the fixed section. Substitute the initial rotation matrix and the initial translation matrix into the iterative closest point ICP algorithm to obtain the accurate rotation matrix and the accurate translation matrix, including step S2. Determine the moving overall segment butting method, which includes two stages. In the first stage, adjust the displacements of the moving overall segment in the three degrees of freedom in the x, y, and z directions and the rotational directions around the x and z axes. In the second stage, adjust the displacement of the moving overall segment in the y direction. Based on the three-degree-of-freedom moving plane position coordinates of the trolley in the global coordinate system obtained in step S1, determine the position of the moving overall segment relative to the trolley. Combine the accurate rotation matrix and the accurate translation matrix obtained in step S2, and obtain the distances that the trolley should move in each degree of freedom through inverse kinematics. Perform a path butting trajectory plan for the trolley through the fifth-order polynomial fitting method, and obtain the trolley motion plan trajectory in step S3. Use a BP neural network to predict the error caused by the deformation at the connection between the trolley and the moving overall segment during path butting adjustment. Train a BP neural network including one input layer, three hidden layers, and one output layer with historical data. Through training prediction, obtain the error values caused by the deformation of the bottom plate member of the moving overall segment due to different speeds, different accelerations, and different weights under the same policy. Thereby, during the trolley path butting adjustment, compensate for the distances that the trolley should move in each degree of freedom obtained in step S3, and recalculate the trolley motion plan trajectory in step S3 to obtain the compensated path butting trajectory in step S4. When controlling and adjusting the trolley, until the actual path butting trajectory of the trolley coincides with the path butting trajectory compensated in step S4, the second laser tracker measures the trolley target ball coordinates on each trolley, and at the same time the laser rangefinder measures the moving distances of each trolley in the three degrees of freedom. Fusion-process the data measured by the second laser tracker and the laser rangefinder, and include step S5 of real-time feedback of the adjustment state of the trolley.

[0005] Furthermore, in step S1, target balls are respectively attached to each key point position and the three-degree-of-freedom moving plane of the trolley including the lower bottom surface, the middle layer surface and the upper layer surface of the trolley. The first laser tracker is used to measure the target ball coordinates on each key point, and the second laser tracker is used to measure the trolley target ball coordinates on each degree-of-freedom moving plane of the trolley. The number of the first laser trackers ensures that all the key point target balls can be measured, and the position of each first laser tracker ensures that the maximum number of key point target balls can be measured. The number of the second laser trackers ensures that the target balls on the three-degree-of-freedom moving planes of all trolleys can be measured, and the position of each second laser tracker ensures that the maximum number of target balls on the three-degree-of-freedom moving planes of the trolleys can be measured.

[0006] The specific method of rough alignment in step S2 is as follows: (101) Calculate the weight of each key point on the moving hull block. Taking each key point as the center respectively, draw a circle with a radius of 1 / 2 of the width of the ship hull block deck. The weight w1 of the i-th key point on the moving hull block is as follows:

Number

Number

Number

Number

Number

[0007] The calculation methods of the exact rotation matrix and the exact translation matrix in step S2 are as follows. (201) Establish the moving ladder mating surface coordinate system and the fixed ladder mating surface coordinate system. Use the g j point as the coordinate origin, and use the direction vector from g j to g k as the x - axis to construct a unit vector.

Number

[0008] The moving total butt joint path alignment trajectory planning in step S3 uses the stepwise posture adjustment method, specifically as follows: (31) Rotate the moving total butt joint by an angle α around the x-axis to move the ship's moving total butt joint to a position horizontal with the x-axis, rotate the moving total butt joint by an angle β around the y-axis to move the ship's moving total butt joint to a position parallel to the y-axis, rotate the moving total butt joint by an angle γ around the z-axis to move the ship's moving total butt joint to a position parallel to the z-axis, and adjust the translation amounts in the two degrees of freedom directions of x and z simultaneously. When the distance threshold set in advance for the interval between the moving block and the fixed block is reached, the translation in the y direction is adjusted to avoid interference during the adjustment of the moving block due to the presence of the block pipe, and the trolley path butting trajectory is planned through the fifth-order polynomial trajectory method.

[0009] Furthermore, step S4 is specifically as follows. (41) Using a BP neural network, predict the butting error caused by the deformation of the plate member of the moving block of ships of various shapes under the same acceleration, the same speed, and different weights. Under the conditions that the assembly environment is the same, the speed and acceleration are the same, and the positions where the trolley supports the block coincide, the control variable is the weight of the ship moving block. By extracting on-site data, predict the butting error caused by moving blocks of different weights, calculate the butting error, obtain the compensation amount in the three-degree-of-freedom directions of each trolley, and add the compensation amount to the trolley path butting trajectory planned in step S3 to adjust the trolley path butting trajectory. (42) Control the weight, trolley position, and assembly environment of the ship moving block to be constant, thereby predicting the block butting deviation under different accelerations and different speeds, compensating the three-degree-of-freedom motion trajectories of each trolley. The BP neural network input layer includes five neurons for selecting weight, speed, acceleration, displacement, and material performance. The hidden layer is set with three layers and includes a plurality of neurons. The output layer is provided with one neuron which is the displacement compensation amount in each direction of each trolley.

[0010] Furthermore, step S5 is specifically as follows. The second laser tracker measures the trolley target ball provided on the three-degree-of-freedom moving plane of the trolley, and at the same time uses a laser rangefinder to measure the three-degree-of-freedom movement of the trolley, and fuses the data measured by the second laser tracker and the laser rangefinder. When it is measured by the second laser tracker that a certain trolley target ball is blocked, the laser rangefinder performs data supplementation. When the power supply of the trolley is cut off or a failure occurs, the trolley feedback measurement device records the current position of the trolley. When the trolley is powered on or a contact failure occurs, the control of the trolley path alignment trajectory is continued until the alignment of the moving section and the fixed section is completed.

[0011] A high-precision measurement and automatic alignment system adapted to the non-redundant sections of a ship according to the present invention, Calibrate the measuring device based on the provided common reference point, calculate the position of the measuring device in the global coordinate system, measure the coordinates of each key point target ball on the fixed section alignment surface and the moving section alignment surface using the measuring device, and at the same time measure the three-degree-of-freedom moving plane position coordinates of the trolley, and a data collection and processing module for converting the coordinates of each key point and the three-degree-of-freedom moving plane position coordinates of the trolley into the global coordinate system by the singular value decomposition method SVD, Taking each key point of the moving assembly as the center and a preset distance as the radius, draw a circle, calculate the weight of each key point of the moving assembly based on the number of pipes in each circle, sort the key points of the moving assembly in ascending order of weight, select any three key points with the largest weight and not collinear to form a plane, find the centroid of this plane, and based on the concept of random combinations, select three key points whose sum of weights other than the above set of combinations is the largest and not collinear to form a plane, find the centroid of this plane, and in turn find the centroid corresponding to the key points of the moving assembly in this way, find the centroid corresponding to the key points on the fixed assembly in the same way, select three pairs of corresponding centroids on the moving assembly and the fixed assembly through the three-point alignment method, and based on the three pairs of corresponding centroids on the selected moving assembly and the fixed assembly, establish the moving assembly mating surface coordinate system and the fixed assembly mating surface coordinate system respectively, calculate the moving assembly mating surface coordinate system and the fixed assembly mating surface coordinate system, obtain the initial rotation matrix and the initial translation matrix, realize the rough alignment between the moving assembly and the fixed assembly, substitute the initial rotation matrix and the initial translation matrix into the iterative closest point ICP algorithm, obtain the accurate rotation matrix and the accurate translation matrix, and the rough alignment and accurate alignment modules for realizing accurate alignment, Determine the position of the trolley relative to the moving assembly. The moving assembly mating method includes two stages. In the first stage, adjust the displacements of the moving assembly in the three degrees of freedom in the x, y, and z directions and the rotation directions around the x and z axes. In the second stage, adjust the displacement of the moving assembly in the y direction. Based on the three-degree-of-freedom moving plane position coordinates of the trolley in the obtained global coordinate system, and the obtained accurate rotation matrix and accurate translation matrix, find the distance that the trolley should move in each degree of freedom through inverse kinematics, and perform a path mating trajectory plan for the trolley through the fifth-order polynomial fitting method based on this distance data. The path mating trajectory planning module, Using a BP neural network, predict the error caused by the deformation of the bottom plate member of the moving platform occurring at the connection between the trolley and the moving platform during the path matching adjustment. Train a BP neural network including one input layer, three hidden layers, and one output layer with historical data, and obtain error values caused by deformations due to different speeds, different accelerations, and different weights under the same policy through training predictions. Thereby, compensate for the movement amount of each degree of freedom of the trolley during the trolley path matching adjustment, and a deformation compensation module for obtaining the path matching trajectory after deformation compensation, When controlling and adjusting the trolley, until the actual path matching trajectory of the trolley coincides with the path matching trajectory after deformation compensation, use a measuring device to measure the trolley target ball coordinates in the three-degree-of-freedom directions of each trolley, and at the same time use the measuring device to measure the planar distance that each trolley moves in the three-degree-of-freedom directions. Fuse the target ball coordinates of each trolley in the three-degree-of-freedom directions and the movement distance data of each trolley in the three-degree-of-freedom directions measured by the measuring device, and include a motion feedback module for real-time feedback of the adjustment state of the trolley.

[0012] The measuring device includes a first laser tracker, a second laser tracker, and a laser distance meter. Target balls are respectively attached to the key point positions and the three-degree-of-freedom movement plane of the trolley including the bottom surface, the middle surface, and the upper surface of the trolley. The first laser tracker is used to measure the target ball coordinates of each key point. The second laser tracker is used to measure the trolley target ball coordinates on the trolley. The laser distance meter is used to measure the planar movement distance of each trolley in the three-degree-of-freedom directions. The number of first laser trackers ensures that all key point target balls can be measured, and the position of each first laser tracker ensures that the most key point target balls can be measured. The number of the second laser trackers ensures that the target balls on the three-degree-of-freedom movement planes of all trolleys can be measured, and the position of each second laser tracker ensures that the target balls on the three-degree-of-freedom movement plane of the trolley with the largest number can be measured.

[0013] The device and equipment of the present invention a memory for storing a computer program executable on a processor, when the computer program operates, it includes a processor for executing steps of a high-precision measurement and automatic alignment method adapted to the crowded sub-assembly of the ship.

Advantages of the Invention

[0014] Beneficial effects: Compared with the prior art, the remarkable technical effects of the technical solution of the present invention are as follows. For the problem that the ship pipes are relatively complex, difficult to align in the alignment process, and have few alignments, a weighting method is adopted to optimize the three-point alignment method, thereby making the finally obtained attitude of the moving sub-assembly more accurate. For the problem that the bottom plate member in the contact area between the ship moving sub-assembly and the trolley is prone to deformation, it is proposed to predict the error caused by the deformation amount of the bottom plate member of the moving sub-assembly under different conditions by a BP neural network, and thereby compensate during the trolley adjustment movement. It is beneficial to reduce the influence of the deformation of the bottom plate member of the moving sub-assembly on the alignment of the sub-assembly, and further improve the alignment accuracy and quality of the two sub-assemblies.

Brief Description of the Drawings

[0015]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0016] Hereinafter, the present invention will be described in detail with reference to the drawings and specific examples.

[0017] As shown in FIGS. 1 and 2, in the high-precision measurement and automatic alignment method applicable to the ship block of the present invention, first, the first to fourth laser trackers are calibrated by measuring the common reference points on the same side. After the calibration is completed, the first laser tracker and the second laser tracker measure the important point target balls of the two blocks, and the third laser tracker and the fourth laser tracker measure the trolley target balls on the four trolleys. Then, based on the important point coordinates, the accurate rotation matrix and the accurate translation matrix between the moving block and the fixed block are obtained by using the weighted and optimized three-point alignment method and the ICP algorithm. At the same time, by measuring the trolley target balls on the four trolleys, the actual position of the trolley before movement is obtained, and the block alignment method is determined. By combining the above accurate rotation matrix, accurate translation matrix and the actual position data of the trolley, the distance that the trolley should theoretically move with three degrees of freedom is obtained through inverse kinematics. Furthermore, a trajectory plan is performed for the trolley using a fifth-order polynomial. The host computer visually determines whether this trajectory plan causes interference between the moving ship block and the fixed block. If it causes interference, different path points are reselected to perform a path plan for the trolley. If there is no interference phenomenon, the BP neural network predicts the error caused by the deformation amount of the bottom plate member of the moving ship block in different cases, thereby obtaining the compensation amount required during the trolley movement. Furthermore, trolley compensation is realized by PLC control. Finally, during the movement, by measuring the trolley target balls by the laser tracker and measuring the trolley with respect to each plane by the laser rangefinder, the difference between the actual movement position and the theoretical movement position of the trolley is obtained, and further feedback is performed on the trolley movement position to control the trolley to realize two-stage adjustment of the moving block, thereby realizing high-precision alignment of the moving block.

[0018] Specifically, it includes the following steps.

[0019] S1. Fix the fixed general section and place it on a trolley with an adjustable moving general section. Provide common reference points that are difficult to deform on the ground on both sides of the fixed general section and the moving general section. Provide key points corresponding to the butting surfaces of the fixed general section and the moving general section. Install a measuring device including a first laser tracker, a second laser tracker, and a laser distance meter, and calibrate the measuring device with the common reference points (i.e., the first laser tracker, the second laser tracker, and the laser distance meter measure the common reference points to calibrate their own positions), calculate the position of the measuring device in the global coordinate system. After calibrating the measuring device, the first laser tracker measures the coordinates of each key point on the butting surfaces of the moving general section and the fixed general section, and the second laser tracker measures the actual position coordinates of each trolley in the three-degree-of-freedom moving plane before adjustment, and convert both the coordinates of each key point and the actual position coordinates of each trolley in the three-degree-of-freedom moving plane before adjustment into the global coordinate system.

[0020] As shown in Fig. 3, 1 - fixed overall section, 2 - moving overall section, 3 - first laser tracker, 4 - second laser tracker, 5 - third laser tracker, 6 - fourth laser tracker, 7 - three - dimensional attitude adjustment trolley, 8 - trolley target ball, 9 - laser rangefinder, 10 - fixed support column, 11 - common reference point target ball, 12 - key point target ball. The fixed overall section 1 is supported by four support columns 10, and the four support columns are symmetric in pairs of two. The moving overall section 2 is provided on four three - dimensional attitude adjustment trolleys. On the three - degree - of - freedom movement planes of the four trolleys, trolley target balls for measuring and feeding back the trolley adjustment distance using a laser tracker are all installed. The three - degree - of - freedom movement planes of the trolley include the bottom surface, the middle layer surface and the upper layer surface of the trolley. Six pairs of corresponding key point sets are provided on the butting surface of the fixed overall section and the butting surface of the moving overall section. At each key point position, a key point target ball 12 for measuring with the first laser tracker later is installed. Eight pairs of common reference points are symmetrically provided on the ground on both sides of the fixed overall section and the moving overall section of the ship, and a common reference point target ball 11 is provided on each common reference point. The arrangement positions of the first and second laser trackers are set so as to measure as many key point target balls as possible using the laser tracker. Then the first and second laser trackers measure the common reference point target balls on the same side as themselves respectively, thereby determining the exact positions of the first and second laser trackers at this time, realizing the position calibration of the first and second laser trackers. At this time, the positions of the first and second laser trackers in the global coordinate system are calculated. After calibration, the first and second laser trackers are used to measure the key point target balls of the fixed overall section and the key point target balls of the moving overall section.Lower the control of the third laser tracker and the fourth laser tracker, place the third and fourth laser trackers at positions where they can measure without blocking the trolley target ball coordinates, fix this position, and the third and fourth laser trackers measure the common reference point target balls on the same side as them respectively. At this time, calibrate the positions of the third and fourth laser trackers, calculate the positions of the third and fourth laser trackers in the global coordinate system, and further measure the trolley target ball coordinates on the four trolleys. Convert the measured target ball coordinates of each key point of the fixed and moving ship sections and the trolley target ball coordinates on each trolley into the global coordinate system by singular value decomposition (SVD) to obtain the target ball coordinates of each key point and each trolley target ball in the global coordinate system.

[0021] As shown in Figure 4, establish the global coordinate system O-XYZ, establish the fixed block coordinate system O1-X1Y1Z1 and the moving block coordinate system O2-X2Y2Z2. Based on the design drawing, select the key points P1~P6 that need to be measured to set the important position coordinates of the design drawing as the fixed block butting surface, and the key points Q1~Q6 that need to be measured to set the moving block butting surface. The key points P1~P6 correspond one-to-one with Q1~Q6, and the positions and distributions of the key points are shown in Figure 5. Provide 16 specific common reference points on the block assembly ground, and the first 8 common point reference points M1~M8 and the latter 8 common reference points M9~M 16It is symmetric with respect to the axial centerline of the ship-fixed block Y1, and the positions and distributions of the common reference points are shown in Fig. 6. Four laser trackers are respectively placed at the placement positions I~IV. The first laser tracker is placed at the placement position I, the second laser tracker is placed at the placement position II, the third laser tracker is placed at the placement position III, and the fourth laser tracker is placed at the placement position IV. The first laser tracker measures the common reference points M1~M8 on the ground, determines the exact position of the first laser tracker by measuring the common reference points, and establishes the coordinate system O3-X3Y3Z3 of the first laser tracker placement position I. The first laser tracker placed at the placement position I measures the key points of the ship-fixed block and the ship-moving block, and obtains the coordinate positions of the measurable key points P1, P2, P3 of the ship-fixed block and the measurable key points Q1, Q2, Q3 of the ship-moving block in the coordinate system O3-X3Y3Z3 of the first laser tracker placement position I. Similarly, the second laser tracker measures the common reference points M9~M 16 to determine the exact position of the second laser tracker and establish the coordinate system O4-X4Y4Z4 of the second laser tracker placement position II. The second laser tracker measures the coordinate positions of the measurable key points P4, P5, P6 of the ship-fixed block and the measurable key points Q4, Q5, Q6 of the ship-moving block in the coordinate system O4-X4Y4Z4 of the second laser tracker placement position II. By fusing and calculating the key point coordinates of the moving block and the fixed block, using the singular value decomposition method (SVD), the key point coordinates of the key points P1, P2, P3, Q1, Q2, Q3 in the coordinate system O3-X3Y3Z3 of the first laser tracker placement position I and the key point coordinates of the key points P4, P5, P6, Q4, Q5, Q6 in the coordinate system O4-X4Y4Z4 of the second laser tracker placement position II are transformed into the global coordinate system to obtain the coordinate positions of each key point in the global coordinate system.

[0022] Similarly, the third laser tracker and the fourth laser tracker respectively measure the common reference points M1~M8 and M9~M on the ground on the same side as them 16Measure it and determine the exact positions of the third laser tracker and the fourth laser tracker by measuring the common reference point. Then, establish the coordinate systems of the third laser tracker placement position III, i.e., O5-X5Y5Z5, and the fourth laser tracker placement position IV, i.e., O6-X6Y6Z6, respectively. The third laser tracker and the fourth laser tracker measure the actual coordinates of the trolley target balls on the trolley on the same side, convert the trolley target ball coordinates obtained by the singular value decomposition (SVD) method into the global coordinate system, and obtain the actual position of the trolley before movement.

[0023] Perform rough alignment on the fixed overall section and the moving overall section, and obtain the initial rotation matrix and the initial translation matrix of the key point sets of the fixed overall section mating surface and the moving overall section mating surface. Then, use the iterative closest point (ICP) algorithm to obtain a more accurate rotation matrix and a more accurate translation matrix. With each key point of the fixed overall section as the center, draw a circle with a preset distance as the radius, calculate the weight of each key point of the fixed overall section based on the number of pipes in each circle, select any three key points with the maximum weight and non-collinearity to form a plane, and find the centroid of this plane. Based on the concept of random combinations, select three key points with the maximum sum of weights other than the above set of combinations and non-collinearity to form a plane, and find the centroid of this plane. In this way, find the centroids corresponding to the key points of the fixed overall section in sequence. Follow the same method to find the centroids corresponding to the key points on the moving overall section, select three pairs of corresponding and non-collinear centroids on the fixed overall section and the moving overall section, establish the coordinate systems of the moving overall section mating surface and the fixed overall section mating surface through the three-point alignment method respectively, calculate the coordinate systems of the moving overall section mating surface and the fixed overall section mating surface constructed through the three-point alignment method, obtain the initial rotation matrix and the initial translation matrix, and realize the rough alignment between the moving overall section and the fixed overall section. This includes substituting the initial rotation matrix and the initial translation matrix into the iterative closest point (ICP) algorithm to obtain the accurate rotation matrix and the accurate translation matrix. This method can accurately align more pipes after the mating of the ship end face is completed, and solves the problems of difficult pipe mating and few accurate mating numbers.

[0024] First, calculate the data measured in step S1. The data calculation performs a fitting search on the measured key point coordinates of the fixed total section and the moving total section. The fitting search first performs a rough alignment on the measured key point sets of the mating surfaces of the fixed total section and the moving total section of the ship through a three-point alignment method, and optimizes the rough alignment method by combining it with pipe mating, including the following.

[0025] (101) With each key point of the moving total section as the center, draw a circle with a preset distance as the radius. Calculate the weight of each key point of the fixed total section based on the number of pipes in each circle. Arrange the key points of the moving total section in ascending order of weight. Select three key points with the largest weight and not collinear to form a plane, and find the centroid of this plane. Then, select three points with the largest weight and not collinear other than the above set of combinations on the ship's moving total section, connect them in order to form a plane, and find the centroid of this plane. In order, find the centroids corresponding to other key points of the moving total section in this way, find the centroids corresponding to each key point on the fixed total section according to the same method, select three pairs of non-collinear corresponding centroids on the moving total section and the fixed total section, establish the coordinate systems of the mating surface of the moving total section and the mating surface of the fixed total section respectively through the three-point alignment method, calculate the coordinate systems of the mating surface of the moving total section and the mating surface of the fixed total section constructed through the three-point alignment method, obtain the initial rotation matrix and the initial translation matrix, and achieve the rough alignment between the moving total section and the fixed total section.

[0026] When butting ship sections, it is necessary to consider the accuracy of pipe butting in the sections. Pipe measurement is to measure manually through tools such as measuring tapes based on the plate member with the closest distance. The measurement reference for ship pipes is generally the strong construction of the ship section butting surface. Since the important measurement points of ship sections are generally at the connection points of the construction, an important point weight is introduced to better control the butting accuracy of the pipes and enable more pipe members to be completely butted. As shown in Figure 7, for the convenience of explaining this method in this embodiment, it is described using 4 out of 6 important points (Q1, Q2, Q4, Q5) on the ship moving section butting surface, and multiple important points more than 4 are analogized according to this method. This method gives weights to each important measurement point by calculating the number of pipes in each circle. The specific content of giving weights is as follows. For Q1, with Q1 as the center, preferably draw a circle R1 with a radius R equal to 1 / 2 of the width of the fixed section deck. The selection of the radius shall not exceed the diagonal distance of the fixed section at most. Similarly, with Q2, Q4, and Q5 as the centers respectively, draw corresponding circles R2, R4, and R5 with a radius of R. The total number of pipes on the ship moving section is M, and the number of pipe centers of the pipes included in the circle R i with the i-th important point on the moving section as the center is N i where, in this embodiment, i = 1, 2, 4, 5. The weight w1 of the i-th important point on the moving section is as follows,

Equation

[0027] (102) The weight set of all important points on the moving section is W = {w1,..., w NIn this case, first select three key points whose sum of weights is the largest and which are not collinear. Connect the three points in pairs to form a planar triangle, calculate the centroid g1 of the triangle, and based on the concept of random combinations, select three key points whose sum of weights is the largest and which are not collinear other than the above set of combinations, form a plane, find the centroid g2 of this plane, and similarly calculate the centroids corresponding to the key points on the moving staircase according to this method.

Number

Number

Number

[0028] (103) Calculate the centroids corresponding to the key points on the fixed staircase according to the same method as above.

Number

Number

Number

[0029] Then, obtain the initial rotation transformation matrix and the initial translation matrix of the key point sets of the moving total stage butting surface and the fixed total stage butting surface, and calculate the exact rotation matrix and the exact translation matrix.

[0030] (201) Establish the moving total stage butting surface coordinate system and the fixed total stage butting surface coordinate system. g j Take the point as the coordinate origin, and use the direction vector from g j to g k as the x-axis to construct a unit vector.

Number

Number

Number

Number

Number

[0031] In this embodiment, after obtaining the weights of each point, first select three important points whose sum of weights is the largest and are not collinear. Connect the three points in sequence to form a plane, calculate the plane centroid g1, and then select three important points whose sum of weights is the largest and are not collinear among those other than the above-mentioned set of combinations on the ship moving block. Connect the three points in sequence to form a plane, calculate the plane centroid g2, and finally select three important points whose sum of weights is the largest and are not collinear among those other than the above-mentioned two sets of combinations on the ship moving block. Connect the three points in sequence to form a plane and calculate the plane centroid g3. Similarly, the corresponding two centroids on the fixed block are

Number

[0032] Take point g1 as the coordinate origin, the direction vector from g1 to g2 as the x-axis, and construct a unit vector,

Number

Number

Number

Number

Number

[0033] S3. First, establish the ship moving block mating method. Different from the mating of other objects, the ship block mating has too large an area, so it needs to be adjusted step by step, and it is necessary to ensure that there is no interference during the mating posture adjustment. Therefore, the whole process of the moving block mating can be divided into two stages. In the first stage, until the moving ship block reaches the target position of the first stage, the rotation around the x, y, and z axes of the three degrees of freedom of the moving block and the translation in the two directions of x and z are adjusted. In the second stage, the translation in the y-axis direction of the moving block is adjusted, and finally the alignment of the moving block is completed. The rotation of the moving block around the x-axis can be realized by the cooperative movement of the four trolleys in the two directions of y and z. Similarly, the rotation of the moving block around the y-axis can be realized by the cooperative movement of the four trolleys in the two directions of x and z. The rotation of the moving block around the z-axis can be realized by the cooperative movement of the four trolleys in the two directions of x and y. The translation of the ship block in the three directions of x, y, and z can be realized by the four trolleys moving simultaneously in the three directions of x, y, and z. After determining the moving block mating method, through the inverse kinematics, the distance that the trolley should move in each degree of freedom during the block mating is calculated by combining the actual position of the trolley before movement obtained in the above step S1 and the accurate rotation matrix and the accurate translation matrix obtained in the above step S2. Finally, the trajectory planning is carried out for the trolley through the fifth-order polynomial fitting method, and the trolley motion planning trajectory is obtained.

[0034] The overall ship moving block trajectory plan uses a step-by-step attitude adjustment method. In the first stage, until the moving block reaches the target position in the first stage, the moving block is rotated by an angle α around the x-axis to move the ship moving block to a position horizontal to the x-axis, rotated by an angle β around the y-axis to move the ship moving block to a position parallel to the y-axis, and rotated by an angle γ around the z-axis to move the ship moving block to a position parallel to the z-axis, and the translation amounts in the two degrees of freedom directions of x and z are adjusted simultaneously. In the second stage, the translation in the y-direction of the moving block is adjusted. Based on the actually measured position of the trolley target ball (i.e., the actual position before the movement of the trolley) and the specific butting method of the moving block, the distances that the trolley should move in three degrees of freedom are obtained through inverse kinematics, and the trolley trajectory is planned through the fifth-order polynomial fitting method. By the two-stage butting method of the moving block, the interference phenomenon caused by the existence of the block pipe is avoided. After the completion of the trajectory plan, it is judged by the visualization function of the upper computer software whether the moving block interferes during the adjustment. If the moving block interferes, the trolley trajectory planning policy is changed by reselecting different path points. If there is no interference, it is determined that the specific trajectory of the trolley is the trolley trajectory currently calculated through the fifth-order polynomial fitting method. Finally, the PLC controls the movement of the trolley. By obtaining the trolley trajectory through the fifth-order polynomial fitting method, the trolley speed and acceleration are made smoother, and problems such as errors and motor damage caused by sudden changes in acceleration are avoided.

[0035] S4. After determining the trolley motion planned trajectory on the premise that the moving girders do not interfere, perform predictive compensation for the error caused by the deformation of the bottom plate member of the moving girder, and compensate for the above error by adjusting the difference in the trolley moving distance and the motion trajectory. Since the masses of different moving girders are different, when the trolley supports the bottom of the moving girder, moving girders of different weights will cause different degrees of deformation of the ship bottom and trolley deformation. The differences in the acceleration and speed adjusted by the trolley will cause slight deformation of the supported part during the adjustment of the moving girder. The compensation of the deformation amount is realized through a BP neural network. First, by controlling the speed and acceleration to be adjusted, the assembly environment, and the material of the bottom plate member of the moving girder, measure, record, and conduct a large amount of training on the deformation amount data of the bottom plate member of the moving girder caused by moving girders of different weights of the ship. Second, measure, record, and conduct a large amount of training on the deformation amount data of the bottom plate member of the moving girder caused by different adjustment speeds and accelerations provided by different types of girders, and further obtain the errors caused by the deformation of the bottom plate members of the moving girders of various shapes of ships under different factors. Output the distance that the trolley should compensate to the neural network to compensate for the deformation of the moving girder plate member. By adjusting the actual motion trajectory of the trolley, compensate for the error caused by the deformation of the bottom plate member of the ship moving girder. Train a BP neural network including one input layer, three hidden layers, and one output layer with historical data. Through training prediction, obtain the error values caused by deformation by different speeds, different accelerations, and different weights respectively under the same policy, thereby compensating the moving amount of each degree of freedom of the trolley before trolley motion adjustment, recalculate the trolley motion trajectory in step S3, and obtain the compensated path matching trajectory.

[0036] Specifically, using a BP neural network, the butting error caused by the deformation of the plate members of the moving gantry of ships of various shapes under the same acceleration, the same speed, and different weights is predicted. Under the conditions that the environment is the same, the speed and acceleration are the same, and the positions where the trolleys support the gantry coincide, the control variable is the weight of the ship moving gantry. By extracting on-site data, the butting error caused by the deformation amount of the bottom plate members due to moving gantries of different weights is predicted. The host computer software calculates the error, obtains the compensation amounts in the three-degree-of-freedom directions of each trolley, adds the adjustment values to the compensation amounts during the process of controlling the trolley movement, thereby reducing the number of butting times and enhancing the butting efficiency of the ship gantry. Similarly, the weight of the ship moving gantry, the trolley position, and the assembly environment are controlled to be constant, thereby predicting the gantry butting deviation under different accelerations and different speeds. Through the calculation of the host computer, the three-degree-of-freedom movement values of each trolley are compensated during the process of controlling the trolley movement. This error compensation system includes a BP neural network input layer, a BP neural network hidden layer, and a BP neural network output layer. The input layer includes five neurons for selecting weight, speed, acceleration, displacement, and material performance. The hidden layer is set with three layers and includes a plurality of neurons. The output layer is provided with one neuron, which is the displacement compensation amount in each direction of each trolley.

[0037] S5. Finally, when adjusting and controlling, use the second laser tracker to measure the trolley target balls in the three-degree-of-freedom directions of each trolley, and at the same time use the laser rangefinder to measure the moving distances in the three-degree-of-freedom directions of each trolley. After the above two devices measure, by processing the measured data, control the adjustment state of the trolley in real time by feedback until the actual adjustment of the trolley coincides with the target adjustment data, that is, until the actual path butting trajectory of the trolley coincides with the path butting trajectory compensated in step S4.

[0038] Specifically, during the movement of the trolley, the laser distance meter measures the movement distance of the trolley in the three-degree-of-freedom plane in real time. After the first-stage adjustment of the total movement stage is completed, the adjustment process is paused. The third and fourth laser trackers of the total stage measure the trolley target balls provided on the three movement planes of the four adjustable trolleys, and further fuse with the data measured by the laser distance meter, take the average value to ensure the accuracy of the data, compare it with the distance that the trolley should theoretically move, add this difference to the adjustment process of the second stage of the total movement stage, and further feedback to the moving position of the trolley. Furthermore, the total movement stage is adjusted to rotate around the x, y, and z axes and translate in the xz direction. It is judged whether the two total stages reach the target position of the first stage. If not, the total movement stage is readjusted to rotate around the x, y, and z axes and translate in the x and z directions until the two total stages reach the target position of the first stage. If so, the total movement stage is adjusted to translate in the y direction.

[0039] The PLC is used to control the trolley. Since the ship's total movement stage has a large mass, large inertia, and high butting requirements, target balls are provided at the target positions in each degree-of-freedom adjustment direction, and the laser tracker is used to measure the target balls of the trolley in each degree-of-freedom direction. Laser distance meters are provided in each degree-of-freedom direction of the trolley to measure the movement distance in each degree of freedom. The data measured by the laser tracker and the data measured by the laser distance meter are processed, and the average value of the two sets of data is taken to reduce the measurement error of the single measurement system. Combining with the distance that the trolley should theoretically move obtained in S4, the error value between the actual movement distance and the theoretical distance of the trolley is calculated, and feedback guidance is provided for the movement of the trolley. The laser tracker measures when a certain trolley target ball is blocked, uses the data measured by the laser distance meter as a supplement to prevent the loss of the measured data, strengthens the robustness of the acquisition of the trolley movement feedback guidance data, and provides data feedback and guidance for each degree-of-freedom stroke that has not moved to the target position.

[0040] The laser tracker has high precision, but there may be a shielding phenomenon during measurement. Therefore, when it is difficult to measure a certain trolley target, the data measured by the laser rangefinder is used for supplementation to ensure the integrity of the data. Also, when a power outage or a failure occurs in the trolley, the trolley feedback measurement device can record the current position. The upper computer performs memory and calculation on the recorded data. After the power supply of the trolley is restored and the failure is eliminated, until the total stage alignment is completed, the motion control, adjustment and detection of the trolley are continued to achieve high precision and automatic alignment of the total stage.

[0041] The laser rangefinder and the laser tracker are used to perform cooperative measurement feedback on the movement of the trolley in each direction, and for situations such as shielding, power outage, and failure, accurate data feedback and recording can be ensured, and it also has good robustness.

[0042] A high-precision measurement and automatic alignment system adapted to the non-redundant total stage of a ship according to the present invention, Calibrate the measurement device based on the provided common reference point, calculate the position of the measurement device in the global coordinate system, use the measurement device to measure the coordinate of each important point target ball on the fixed total stage alignment surface and the moving total stage alignment surface, and at the same time measure the three-degree-of-freedom moving plane position coordinate of the trolley, and a data collection and processing module for converting the coordinate of each important point and the three-degree-of-freedom moving plane position coordinate of the trolley into the global coordinate system by the singular value decomposition method SVD, Taking each key point of the moving assembly as the center and a preset distance as the radius, draw a circle, calculate the weight of each key point of the moving assembly based on the number of pipes in each circle, sort each key point of the moving assembly in ascending order of weight, select any three key points with the largest weight and not collinear to form a plane, find the centroid of this plane, and based on the concept of random combinations, select three key points with the largest sum of weights other than the above set of combinations and not collinear to form a plane, find the centroid of this plane, and in turn find the centroid corresponding to the key points of the moving assembly in this way, find the centroid corresponding to the key points on the fixed assembly according to the same method, select three pairs of corresponding centroids on the moving assembly and the fixed assembly through the three-point alignment method, and based on the three pairs of corresponding centroids on the selected moving assembly and the fixed assembly, establish the moving assembly mating surface coordinate system and the fixed assembly mating surface coordinate system respectively, calculate the moving assembly mating surface coordinate system and the fixed assembly mating surface coordinate system, obtain the initial rotation matrix and the initial translation matrix, realize the rough alignment between the moving assembly and the fixed assembly, substitute the initial rotation matrix and the initial translation matrix into the iterative closest point ICP algorithm to obtain the accurate rotation matrix and the accurate translation matrix, and the rough alignment and accurate alignment modules for realizing accurate alignment, Determine the position of the trolley relative to the moving assembly. The moving assembly mating method includes two stages. In the first stage, adjust the displacements of the moving assembly in the three degrees of freedom of x, y, and z directions and the rotation directions around the x and z axes. In the second stage, adjust the displacement of the moving assembly in the y direction. Based on the three-degree-of-freedom moving plane position coordinates of the trolley in the obtained global coordinate system, and the obtained accurate rotation matrix and accurate translation matrix, find the distance that the trolley should move in each degree of freedom through inverse kinematics, and perform a path mating trajectory plan for the trolley through a fifth-order polynomial fitting method based on this distance data. The path mating trajectory planning module, Using a BP neural network, predict the error caused by the deformation of the bottom plate member of the moving staircase at the connection between the trolley and the moving staircase during path alignment adjustment. Train a BP neural network including one input layer, three hidden layers, and one output layer with historical data, and obtain error values caused by deformations due to different speeds, different accelerations, and different weights under the same policy through training prediction. Thereby, compensate for the movement amount of each degree of freedom of the trolley during trolley path alignment adjustment, and a deformation compensation module for obtaining the path alignment trajectory after deformation compensation, When controlling and adjusting the trolley, until the actual path alignment trajectory of the trolley coincides with the path alignment trajectory after deformation compensation, use a measuring device to measure the trolley target ball coordinates in the three-degree-of-freedom direction of each trolley, and at the same time use the measuring device to measure the plane distance that each trolley moves in the three-degree-of-freedom direction. Fuse the target ball coordinates in the three-degree-of-freedom direction of each trolley measured by the measuring device and the movement distance data in the three-degree-of-freedom direction of each trolley, and include a motion feedback module for real-time feedback of the adjustment state of the trolley.

[0043] The measuring device includes a first laser tracker, a second laser tracker, and a laser rangefinder. Attach target balls to the key point positions and the three-degree-of-freedom movement plane of the trolley including the bottom surface, middle layer surface, and upper layer surface of the trolley respectively. The first laser tracker is used to measure the target ball coordinates of each key point. The second laser tracker is used to measure the trolley target ball coordinates on the trolley. The laser rangefinder is used to measure the plane movement distance in the three-degree-of-freedom direction of each trolley. The number of the first laser trackers ensures that all key point target balls can be measured, and the position of each first laser tracker ensures that the most key point target balls can be measured. The number of the second laser trackers ensures that the target balls on the three-degree-of-freedom movement planes of all trolleys can be measured, and the position of each second laser tracker ensures that the target balls on the three-degree-of-freedom movement plane of the trolley with the largest number can be measured.

[0044] An apparatus and equipment of the present invention, a memory for storing a computer program executable on a processor, and a processor for executing steps of a high-precision measurement and automatic alignment method adapted to the crowded ship block when the computer program operates.

Claims

1. A high-precision measurement and automatic alignment method adapted to a ship's crowded block, comprising: fixing and providing a fixed block, placing a movable block on an adjustable trolley, providing common reference points on the ground on both sides of the fixed block and the movable block, providing corresponding key points on the fixed block mating surface and the movable block mating surface, installing a measuring device including a first laser tracker, a second laser tracker and a laser rangefinder, calibrating the measuring device by the common reference points, calculating the position of the measuring device in the global coordinate system, measuring the coordinates of each key point using the first laser tracker, measuring the three-degree-of-freedom moving plane position coordinates of the trolley using the second laser tracker, and converting both the coordinates of each key point and the three-degree-of-freedom moving plane position coordinates of the trolley into the global coordinate system in step S1; performing rough alignment on the fixed block and the movable block, and obtaining an initial rotation matrix and an initial translation matrix of the key point set on the fixed block mating surface and the movable block mating surface, and further obtaining an accurate rotation matrix and an accurate translation matrix using the iterative closest point ICP algorithm in step S2, which includes drawing a circle with each key point of the movable block as the center and a preset distance as the radius, calculating the weight of each key point of the movable block based on the number of pipes in each circle, selecting three key points with the maximum sum of weights and not collinear to form a plane, finding the centroid of this plane, based on the concept of random combinations, selecting three key points with the maximum sum of weights and not collinear other than the above set of combinations to form a plane, finding the centroid of this plane, and sequentially finding the centroids corresponding to the key points of the movable block in this way, finding the centroids corresponding to the key points on the fixed block in the same way, selecting three pairs of corresponding and non-collinear centroids on the fixed block and the movable block, establishing the movable block mating surface coordinate system and the fixed block mating surface coordinate system respectively through the three-point alignment method, calculating the two coordinate systems, obtaining the initial rotation matrix and the initial translation matrix, and realizing the rough alignment between the movable block and the fixed block, and substituting the initial rotation matrix and the initial translation matrix into the iterative closest point ICP algorithm to obtain the accurate rotation matrix and the accurate translation matrix in step S2; Determine the moving total segment butting method, which includes two stages. In the first stage, adjust the displacements of the moving total segment in the three degrees of freedom in the x, y, and z directions and the rotational directions around the x and z axes. In the second stage, adjust the displacement of the moving total segment in the y direction. Based on the three-degree-of-freedom moving plane position coordinates of the trolley in the global coordinate system obtained in step S1, determine the position of the moving total segment relative to the trolley. Combine the accurate rotation matrix and the accurate translation matrix obtained in step S2, and obtain the distances that the trolley should move in each degree of freedom through inverse kinematics. Perform a path butting trajectory plan for the trolley through the fifth-order polynomial fitting method, and obtain the trolley motion plan trajectory in step S3. Use a BP neural network to predict the error caused by the deformation at the connection between the trolley and the moving total segment during the path butting adjustment. Train a BP neural network including one input layer, three hidden layers, and one output layer with historical data. Obtain the error values caused by the deformation of the bottom plate member of the moving total segment under different speeds, different accelerations, and different weights under the same policy through training prediction. Thereby, during the trolley path butting adjustment, compensate for the distances that the trolley should move in each degree of freedom obtained in step S3, and recalculate the trolley motion plan trajectory in step S3 to obtain the compensated path butting trajectory in step S4. When controlling and adjusting the trolley, until the actual path butting trajectory of the trolley coincides with the path butting trajectory compensated in step S4, the second laser tracker measures the trolley target ball coordinates on each trolley, and at the same time the laser rangefinder measures the moving distances of each trolley in the three degrees of freedom directions. Fuse the data measured by the second laser tracker and the laser rangefinder, and feedback the adjustment state of the trolley in real time in step S5. A high-precision measurement and automatic butting method adapted to the total segment of a ship without margin, characterized by the above.

2. In step S1, target balls are respectively attached to each key point position and the three-degree-of-freedom moving plane of the trolley including the lower bottom surface, the middle layer surface and the upper layer surface of the trolley. The first laser tracker is used to measure the coordinates of the target balls on each key point, and the second laser tracker is used to measure the coordinates of the trolley target balls on each degree-of-freedom moving plane of the trolley. The number of the first laser trackers ensures that all the key point target balls can be measured, and the position of each first laser tracker ensures that the most key point target balls can be measured. The number of the second laser trackers ensures that the target balls on the three-degree-of-freedom moving planes of all trolleys can be measured, and the position of each second laser tracker ensures that the most target balls on the three-degree-of-freedom moving planes of the trolleys can be measured. The high-precision measurement and automatic alignment method applicable to the ship's full-section without margin described in claim 1 is characterized by this.

3. The specific method of rough alignment in step S2 is as follows. (101) Calculate the weights of each key point on the moving sub-assembly. Taking each key point as the center respectively, draw a circle with a radius of 1 / 2 of the width of the ship sub-assembly deck. The weight w of the i-th key point on the moving sub-assembly １ is as follows: 【Number 29】 However, M is the total number of moving total-stage pipes, and N １ is the number of pipe centers included in the circle R with the i-th key point on the moving total stage as the center, where i is the i-th key point number on the moving total stage, and the weight set of all key points on the moving total stage is W = {w1,..., w １}, and N is the total number of key points on the moving total stage, Ｎ} and N is the total number of key points on the moving total stage, For the set of weights \(W = \{w_1,\ldots,w\}\) of all key points on the moving sub - segment Ｎ}, first select three key points whose sum of weights is the largest and are not collinear. Connect the three points in pairs to form a planar triangle, and calculate the centroid \(g\) of the triangle １ . Based on the concept of random combinations, select three key points whose sum of weights is the largest and are not collinear, other than the above combination. Connect the three points in pairs to form a planar triangle, and calculate the centroid \(g\) of this triangle ２ . Similarly, according to this method, calculate the centroids corresponding to all possible combinations of three key points among all key points on the moving sub - segment 【Count 30】 Calculate, where H is the number of sets of 3 collinear key points selected. (103) According to the same method, calculate the center of mass corresponding to the key points on the fixed full-section 【Number 31】 and perform alignment and matching by the three-point alignment method for 3 non-collinear centers of mass 【Number 32】 on the moving full-section and the corresponding 3 centers of mass 【Number 33】 on the fixed full-section to achieve rough alignment between the moving full-section and the fixed full-section. The high-precision measurement and automatic alignment method applicable to the ship's full-section without margin described in claim 1 is characterized by this.

4. The calculation method of the exact rotation matrix and the exact translation matrix in step S2 is as follows. (201) Establish the moving full-section mating surface coordinate system and the fixed full-section mating surface coordinate system. g ｊ Taking the point as the origin of coordinates, with the direction vector from g ｊ to g ｋ as the x-axis, construct a unit vector, 【Number 34】 And g ｊ from g ｌ determine the y-axis by the product of the direction vector to and the x-axis direction vector, construct the y-axis unit vector, 【Number 35】 Finally, e １ and e ２ are used to determine the z-axis direction, and its unit vector is as follows: e ３ = e １ × e ２ The moving total segment butting surface coordinate system is O ｇ -e １ , e ２ , e ３ and is Similarly, determine and the fixed full-section mating surface coordinate system 【Number 36】 and it is (202) Calculate the initial rotation matrix and the initial translation matrix. By aligning the measured data using the three-point alignment method, the initial rotation transformation matrix R of the set of key points on the total moving butt joint surface and the total fixed butt joint surface is obtained. ０ It is expressed as follows: 【No. 37】 Initial rotation matrix R ０ Based on this, the initial translation matrix t ０ is calculated as follows: 【Number 38】 (203) Initial rotation matrix R ０ and initial translation matrix t ０ are substituted into the iterative closest point ICP algorithm, and through iterative calculations, an accurate rotation matrix R and an accurate translation matrix t are obtained. The high-precision measurement and automatic butt joint method according to claim 3, which is characterized in that it is adapted to the full block of the ship without margin.

5. The moving full-section path mating trajectory planning in step S3 uses the step-by-step posture adjustment method, and specifically is as follows. (31) Rotate the moving overall section by an angle α around the x-axis to move the ship's moving overall section to a position horizontal to the x-axis, rotate the moving overall section by an angle β around the y-axis to move the ship's moving overall section to a position parallel to the y-axis, rotate the moving overall section by an angle γ around the z-axis to move the ship's moving overall section to a position parallel to the z-axis, and adjust so that the translational amounts in the two degrees of freedom directions of x and z are adjusted simultaneously, (32) When the distance between the moving overall section and the fixed overall section reaches a preset distance threshold, adjust the translation in the y-direction to avoid interference during the adjustment of the moving overall section due to the presence of the overall section pipes, and plan the trolley path butting trajectory through the fifth-order polynomial trajectory method. A high-precision measurement and automatic butting method applicable to the non-redundant overall section of a ship according to claim 1, characterized in that.

6. Step S4 is specifically as follows: (41) Using a BP neural network, predict the butting error caused by the deformation of the plate members of the moving overall section of ships of various shapes under the same acceleration, the same speed, and different weights. Under the conditions that the assembly environment is the same, the speed and acceleration are the same, and the positions where the trolleys support the overall section coincide, the control variable is the weight of the ship's moving overall section. By extracting on-site data, predict the butting error caused by moving overall sections of different weights, calculate the butting error, obtain the compensation amounts in the three degrees of freedom directions of each trolley, and add the compensation amounts to the trolley path butting trajectory planned in step S3 to adjust the trolley path butting trajectory. (42) Control the weight, trolley position, and assembly environment of the ship's moving overall section to be constant, thereby predicting the overall section butting deviation under different accelerations and different speeds, compensating the three-degree-of-freedom motion trajectories of each trolley. The BP neural network input layer includes five neurons for selecting weight, speed, acceleration, displacement, and material performance. The hidden layer is set in three layers and includes a plurality of neurons. The output layer is provided with one neuron that is the displacement compensation amount in each direction of each trolley. A high-precision measurement and automatic butting method applicable to the non-redundant overall section of a ship according to claim 1, characterized in that.

7. Step S5 is specifically as follows: The second laser tracker measures a trolley target ball provided on the three-degree-of-freedom movement plane of the trolley, and at the same time uses a laser distance meter to measure the three-degree-of-freedom movement of the trolley, and fuses the data measured by the second laser tracker and the laser distance meter. When it is measured by the second laser tracker that a certain trolley target ball is shielded, the laser distance meter performs data supplementation. When the power supply of the trolley is cut off or a failure occurs, the trolley feedback measurement device records the current position of the trolley. When the trolley is powered on or a contact failure occurs, the control of the trolley path alignment trajectory is continued until the alignment of the moving section and the fixed section is completed. The high-precision measurement and automatic alignment method according to claim 1, which is applicable to the non-redundant sections of a ship, characterized in that.

8. A high-precision measurement and automatic alignment system applicable to non-redundant sections of a ship, comprising: Calibrating a measuring device based on a provided common reference point, calculating the position of the measuring device in the global coordinate system, using the measuring device to measure the coordinates of each key point target ball on the fixed-section alignment surface and the moving-section alignment surface, and at the same time measuring the three-degree-of-freedom moving plane position coordinates of the trolley, and a data collection and processing module for converting both the coordinates of each key point and the three-degree-of-freedom moving plane position coordinates of the trolley into the global coordinate system by the singular value decomposition method SVD; Taking each key point of the moving sub - section as the center and a preset distance as the radius to draw a circle, calculating the weight of each key point of the moving sub - section based on the number of pipes in each circle, sorting each key point of the moving sub - section in ascending order of weight, selecting any three key points with the largest weight and not collinear to form a plane, finding the centroid of this plane, based on the concept of random combinations, selecting three key points with the largest sum of weights other than the above - mentioned set of combinations and not collinear to form a plane, finding the centroid of this plane, and in turn finding the centroids corresponding to the key points of the moving sub - section in this way, finding the centroids corresponding to the key points on the fixed sub - section according to the same method, selecting three pairs of corresponding centroids on the moving sub - section and the fixed sub - section through the three - point alignment method, establishing the moving sub - section mating surface coordinate system and the fixed sub - section mating surface coordinate system respectively based on the three pairs of corresponding centroids on the selected moving sub - section and the fixed sub - section, calculating the moving sub - section mating surface coordinate system and the fixed sub - section mating surface coordinate system, obtaining the initial rotation matrix and the initial translation matrix, realizing the rough alignment between the moving sub - section and the fixed sub - section, substituting the initial rotation matrix and the initial translation matrix into the iterative closest point ICP algorithm to obtain the accurate rotation matrix and the accurate translation matrix, a rough alignment and accurate alignment module for realizing accurate alignment, Determining the position of the trolley relative to the moving sub - section, the moving sub - section mating method includes two stages. In the first stage, the displacements of the moving sub - section in the three degrees of freedom in the x, y, and z directions and the rotation directions around the x and z axes are adjusted. In the second stage, the displacement of the moving sub - section in the y direction is adjusted. Based on the three - degree - of - freedom moving plane position coordinates of the trolley in the obtained global coordinate system, and the obtained accurate rotation matrix and accurate translation matrix, the distances that the trolley should move in each degree of freedom are obtained through inverse kinematics, and based on this distance data, a path mating trajectory planning is carried out for the trolley through a fifth - order polynomial fitting method. A path mating trajectory planning module, Using a BP neural network, predict the error caused by the deformation of the bottom plate member of the moving bulkhead occurring at the connection between the trolley and the moving bulkhead during path butting adjustment. Train a BP neural network including one input layer, three hidden layers, and one output layer with historical data, and obtain error values caused by deformations due to different speeds, different accelerations, and different weights under the same policy through training prediction. Thereby, compensate for the movement amounts of each degree of freedom of the trolley during trolley path butting adjustment, and a deformation compensation module for obtaining a path butting trajectory after deformation compensation, When controlling and adjusting the trolley, until the actual path butting trajectory of the trolley coincides with the path butting trajectory after deformation compensation, use a measuring device to measure the trolley target ball coordinates in the three degrees of freedom directions of each trolley, and at the same time use the measuring device to measure the plane distance that each trolley moves in the three degrees of freedom directions. Fuse the target ball coordinates in the three degrees of freedom directions of each trolley measured by the measuring device and the movement distance data in the three degrees of freedom directions of each trolley, and include a motion feedback module for real-time feedback of the adjustment state of the trolley. A high-precision measurement and automatic butting system adapted to the non-redundant bulkhead of a ship, characterized in that.

9. The measuring device includes a first laser tracker, a second laser tracker, and a laser distance meter. Target balls are respectively attached to the key point positions and the three-degree-of-freedom movement plane of the trolley including the bottom surface, the middle layer surface, and the upper layer surface of the trolley. The first laser tracker is used to measure the target ball coordinates of each key point. The second laser tracker is used to measure the trolley target ball coordinates on the trolley. The laser distance meter is used to measure the plane movement distance in the three degrees of freedom directions of each trolley. The number of first laser trackers ensures that all key point target balls can be measured, and the position of each first laser tracker ensures that the most key point target balls can be measured. The number of the second laser trackers ensures that the target balls on the three-degree-of-freedom movement plane of all trolleys can be measured, and the position of each second laser tracker ensures that the target balls on the three-degree-of-freedom movement plane of the trolley with the largest number can be measured. The high-precision measurement and automatic alignment system according to claim 8, which is applicable to the ship's crowded block without margin.

10. An apparatus and equipment, a memory for storing a computer program executable on a processor, a processor for executing steps of the high-precision measurement and automatic alignment method according to any one of claims 1-7 when the computer program operates, and an apparatus and equipment characterized by the above.

Images