Ship section butt joint attitude adjustment method, system, medium and equipment

Abstract

The application provides a ship section butt joint attitude adjustment method, system, medium and equipment. The attitude adjustment method proposes the concept of rigid fixed point, so that the butt joint ship section type is expanded from small rigid partition to large flexible section, and is not limited to small rigid section. The pose judgment point is used as the reference data of the adjustment instruction, the butt joint calculation principle does not need to be understood, only the real-time measurement point cloud is brought into the method, the pose judgment point cloud real-time parameter with a limited number of pose judgment points can be calculated, the pose deviation is calculated by comparing the pose judgment point cloud after butt joint, and the ship section pose adjustment can be performed according to the pose deviation. Meanwhile, the application proposes that the measurement points are matched with the feature points in the form of voting, and a quantifiable butt joint measurement data quality evaluation method is also proposed. The application caters to the adjustment habit and intuition of the on-site construction personnel, reduces the threshold of the butt joint process, reduces the time and labor consumption of the butt joint process, and improves the shipbuilding efficiency.

Description

Technical Field

[0001] This application relates to the field of shipbuilding, and more specifically, to a method, system, medium, and equipment for docking and adjusting ship sections. Background Technology

[0002] A ship section refers to a ring-shaped section of the hull formed by dividing the main hull laterally along the length of the ship according to the technological requirements of hull construction and the characteristics of the hull structure.

[0003] In the current mainstream modular shipbuilding technology, modular docking is a key link in modular construction. It requires the use of wide-area measurement equipment to track and measure the modular sections. Combined with the judgment and experience of on-site construction personnel, docking vehicles are generally used to issue adjustment and docking instructions to the modular sections.

[0004] In conventional assembly docking technology, to ensure smooth docking of ship assemblies with other assemblies as required, the attitude of the ship assemblies needs continuous adjustment. To obtain the attitude adjustment parameters, workers must constantly measure parameters at various points on the ship assemblies and perform complex calculations to arrive at the docking attitude adjustment parameters. The entire attitude adjustment process is cumbersome, the data collected and judged during the process lacks effective precision, the process of obtaining the attitude adjustment parameters is difficult and inconvenient to use, consequently increasing the time and manpower required for assembly docking and reducing shipbuilding efficiency. If the attitude adjustment parameters are inappropriate, it may further cause significant stress changes in the ship assemblies, potentially damaging the structure of the assemblies. Summary of the Invention

[0005] The purpose of this application is to provide a method, system, medium, and equipment for docking and adjusting ship sections, which caters to the adjustment habits and intuition of on-site construction personnel, lowers the threshold of the docking process, reduces the time and manpower consumption of the docking process, and improves shipbuilding efficiency.

[0006] Firstly, a method for docking and adjusting the attitude of ship sections is provided, characterized by the following steps:

[0007] S10. Divide the ship's main sections into rigid zones, and establish the rigid stationary point cloud O of the ship's main sections and the measurement point cloud M of each rigid zone. j The simulation yielded a point cloud M for each measurement point. j The mapping matrix T from the total cloud of measurement points M to the rigid fixed point cloud O. o A rigid fixed point cloud O contains multiple rigid points, and the relative positions of these rigid points vary within a predetermined range; M j Represents the measurement point cloud of the j-th rigid partition;

[0008] S20. Establish the pose determination point cloud P of the ship section. This is done by inputting a set of measurement point clouds M0 and pose determination point clouds P0 obtained at the same time and in the same attitude. Then, the mapping relationship matrix T is used to determine the pose determination point cloud P. o Calculate a rigid, stationary point cloud O0, and then calculate the mapping matrix T from the rigid, stationary point cloud O0 to the pose of the point cloud P0. p ;

[0009] S30. Simulate docking of the ship sections. After docking, extract multiple feature points from each rigid partition to form a feature point cloud N for each partition. j Each feature point cloud N j The total feature point cloud N is formed by mapping relationship matrix T o Obtain the rigid fixed point cloud O after docking n , from rigid fixed point cloud O n Through the mapping relationship matrix T p The pose determination point cloud P after docking is obtained n ;

[0010] S40. Before the docking of the ship section is completed, based on the real-time measurement point cloud M, find the measurement point with the highest matching degree with the feature point in the feature point cloud N in the measurement point cloud M, and assign the measurement point with the highest matching degree and the feature point the same number.

[0011] S50. Perform homogeneous transformation matching from the total cloud of measurement points M to the total cloud of feature points N. The homogeneous transformation matrix is ​​T. Evaluate the reliability of measurement points with the same index as the feature points, and then determine whether the reliability of the total cloud of measurement points M meets the predetermined requirements. If the predetermined requirements are not met, repeat steps S40 and S50. If the predetermined requirements are met, continue to step S60.

[0012] S60. Multiply the total feature point cloud N by the inverse of the homogeneous transformation matrix T to obtain the real-time total feature point cloud N of the ship section under the current attitude. t The real-time feature point total cloud N t With mapping relation matrix T o Multiplication yields the real-time rigid fixed point cloud O. t , to transform real-time rigid fixed point cloud O t With mapping relation matrix T p Multiplication yields the real-time pose determination point cloud P. t Calculate the real-time pose determination point cloud P t Pole cloud P after docking n The pose deviation ΔP; the pose deviation includes the positional deviation of each real-time pose judgment point and the pose judgment point after docking in three-dimensional space.

[0013] S70. Convert the position deviation ΔP into an attitude adjustment command; send the attitude adjustment command to the docking vehicle used for docking the ship section so that the docking vehicle can make corresponding actions, or provide the attitude adjustment command to the staff for reference to adjust the attitude of the ship section.

[0014] In one feasible approach, the pose determination point is a solid point on the ship section, including at least the bow and stern bottom deflection points and the four horizontal points at the top corners of the ship section.

[0015] In one feasible scheme, the rigid partitions are symmetrically distributed along the mid-longitudinal section of the ship section, and the measurement points in the symmetrical rigid partitions are also configured to be symmetrically distributed along the mid-longitudinal section of the ship section.

[0016] In one feasible approach, all the fixed points of the rigid fixed point cloud are configured within the mid-longitudinal section of the ship section.

[0017] In one feasible embodiment, step S40 includes the following steps:

[0018] S41. Calculate the distance feature set of each feature point relative to the other feature points;

[0019] S42. Measure the coordinate data of the measurement points of the ship section in real time, and calculate the distance feature set of each measurement point relative to the other measurement points according to the rules in step S41.

[0020] S43. Establish a similarity judgment voting mechanism between measurement points and feature points: If there are distance features in the distance feature set between a feature point and a measurement point that differ within a predetermined difference, then the similarity score between the two is increased by one; calculate the matching degree between each measurement point and each feature point according to the aforementioned rules.

[0021] S44. Assign the same number to the feature point and the measurement point with the highest matching degree.

[0022] In one feasible embodiment, step S50 includes the following steps:

[0023] S511. In each rigid partition, select no fewer than three pairs of measurement points and feature points with the highest similarity. The measurement points selected in each rigid partition form the temporary measurement point total cloud M of the current partition. tem The feature points selected from each rigid partition form a temporary feature point cloud N. tem Calculate the total cloud M from the temporary measurement points in each partition. tem Total cloud N to temporary feature point tem Temporary homogeneous transformation matrix T j ;T j This represents the temporary homogeneous transformation matrix corresponding to the j-th rigid partition;

[0024] S512, transfer each measurement point cloud M j Use the corresponding temporary homogeneous transformation matrix T j Perform the transformation;

[0025] S513. Set a matching error threshold between the measurement points and feature points. Match the transformed measurement points with the feature points. If the matching error is greater than the matching error threshold, exclude the corresponding measurement points and feature points to obtain the corrected measurement point cloud M for each rigid partition. a Total cloud of corrected feature points N a ;

[0026] S514, Calculate the point cloud M of each rigid partition from the calibration measurement point cloud. a To correct the feature point cloud N a The corrected homogeneous transformation matrix T a ;T a This represents the correction homogeneous transformation matrix corresponding to the a-th rigid partition;

[0027] S515. Calculate the number K of measurement points greater than the matching error threshold within each rigid partition. The specific expression for K is:

[0028]

[0029] Where q represents the measured point cloud M j The number of measurement points in the middle;

[0030] α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points;

[0031] m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ;

[0032] n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ;

[0033] Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points;

[0034] The term indicates that its internal values ​​are rounded down;

[0035] S516. Determine whether the ratio of the K value in each rigid partition to the number of measurement points in the corresponding rigid partition is less than 0.3; if it is less than 0.3, then the reliability of the total cloud M of the measurement points meets the predetermined requirements, and the correction homogeneous transformation matrix T corresponding to all rigid partitions is determined. a Together they form a homogeneous transformation matrix T; if there is a case greater than 0.3, it indicates that the reliability of the total cloud M of the measurement points does not meet the predetermined requirements, and steps S40 and S50 need to be repeated.

[0036] In one feasible embodiment, step S50 includes the following steps:

[0037] S521, Set the measurement point cloud M in each rigid partition. j To feature point cloud N j The partition homogeneous transformation matrix is ​​T a , where a = j;

[0038] S522. Calculate the homogeneous transformation matrix of the rigid partition point cloud when the error value function of each rigid partition reaches its minimum value. a The error value function is:

[0039]

[0040] Where q represents the measured point cloud M j The number of measurement points in the middle;

[0041] α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points;

[0042] m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ;

[0043] n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ;

[0044] Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points;

[0045] The term indicates that its internal values ​​are rounded down;

[0046] S523. Calculate the number K of measurement points greater than the matching error threshold within each rigid partition. The specific expression for K is:

[0047]

[0048] Among them, T a This refers to the partition homogeneous transformation matrix calculated in step S522;

[0049] q represents the measured point cloud M j The number of measurement points in the middle;

[0050] α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points;

[0051] m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ;

[0052] n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ;

[0053] Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points;

[0054] The term indicates that its internal values ​​are rounded down;

[0055] S524. Determine whether the ratio of the K value in each rigid partition to the number of measurement points in the corresponding rigid partition is less than 0.3, and whether the minimum value of the error value function is less than or equal to a predetermined value; if they are all less than 0.3 and the minimum value of the error value function is less than or equal to the predetermined value, then the reliability of the total cloud M of the measurement points meets the predetermined requirements, and the partition homogeneous transformation matrix T corresponding to all rigid partitions is determined. a Together they form a homogeneous transformation matrix T; otherwise, it indicates that the reliability of the total cloud M at the measurement point does not meet the predetermined requirements, and steps S40 and S50 need to be repeated.

[0056] According to a second aspect of this application, a docking and attitude adjustment system for ship sections is also provided, including a point cloud establishment module, a mapping relationship calculation module, a simulated docking module, a matching module, a reliability judgment module, a deviation calculation module, and a conversion module.

[0057] The point cloud creation module is used to divide the ship section into rigid zones, and to create the rigid stationary point cloud O of the ship section and the measurement point cloud M of each rigid zone. j This method establishes a point cloud P for determining the pose of the ship section; it is also used to extract multiple feature points in each rigid partition after simulated docking of the ship section to form a feature point cloud N for each partition. j The mapping relationship calculation module is used to simulate and obtain the mapping relationship from each measured point cloud M. j The mapping matrix T from the total cloud of measurement points M to the rigid fixed point cloud O. o It is also used to input a set of total measurement point cloud M0 and pose determination point cloud P0 obtained at the same time and in the same posture, and then to use the mapping relationship matrix T o Calculate a rigid, stationary point cloud O0, and then calculate the mapping matrix T from the rigid, stationary point cloud O0 to the pose of the point cloud P0. p The simulated docking module is used to simulate the docking of ship sections, and will be generated by each feature point cloud N. j The total feature point cloud N is formed by mapping relationship matrix T o Obtain the rigid fixed point cloud O after docking n , from rigid fixed point cloud O n Through the mapping relationship matrix T p The pose determination point cloud P after docking is obtained n The matching module is used to find the measurement point with the highest matching degree with the feature point in the feature point total cloud N in the measurement point total cloud M based on the real-time measurement point total cloud M, and assign the measurement point with the highest matching degree and the feature point the same index. The reliability judgment module is used to perform homogeneous transformation matching from the measurement point total cloud M to the feature point total cloud N, with homogeneous transformation matrix T, to evaluate the reliability of the measurement point with the same index as the feature point, and then determine whether the reliability of the measurement point total cloud M meets the predetermined requirements. The deviation calculation module is used to multiply the feature point total cloud N by the inverse matrix of the homogeneous transformation matrix T to obtain the real-time feature point total cloud N under the current attitude of the ship section. t The real-time feature point total cloud N t With mapping relation matrix T o Multiplication yields the real-time rigid fixed point cloud O. t , to transform real-time rigid fixed point cloud O t With mapping relation matrix T p Multiplication yields the real-time pose determination point cloud P. t Calculate the real-time pose determination point cloud P t Pole cloud P after docking n The pose deviation ΔP is converted into a pose adjustment command by a conversion module.

[0058] According to a third aspect of this application, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the ship section docking and attitude adjustment method.

[0059] According to a fourth aspect of this application, a computer-readable storage medium is also provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of the docking and attitude adjustment method for ship sections.

[0060] Compared with the prior art, the beneficial effects of this application are as follows:

[0061] This application proposes the concept of rigid fixed points, expanding the types of ship sections that can be docked from small rigid sections to large flexible sections, no longer limited to small rigid sections. Using pose judgment points as reference data for adjustment commands, users do not need to understand the specific docking calculation principles; they only need to obtain the real-time measurement point cloud. They can then calculate the real-time parameters of the pose judgment point cloud with a finite number of pose judgment points using the aforementioned method, and calculate the pose deviation by comparing it with the pose judgment point cloud after docking. Workers can then adjust the pose of the ship section based on the pose deviation, catering to the adjustment habits and intuition of on-site construction personnel, lowering the threshold of the docking process, reducing the time and manpower consumption during docking, and improving shipbuilding efficiency.

[0062] Meanwhile, this application proposes a method for pairing measurement points with feature points through voting, which solves the problems of measurement point loss and disordered order during the measurement process. In addition, the voting score can also be used as a coarse matching criterion for subsequent quality assessment of measurement points.

[0063] Furthermore, this application proposes a quantifiable method for evaluating the quality of docking measurement data. Users can quickly assess the accuracy and quality of measurement points during docking, thereby ensuring that the data obtained from the measurement points meet the requirements. Consequently, the attitude deviation and other related parameters obtained later can meet the actual attitude adjustment needs, thus ensuring that the attitude adjustment parameters are appropriate. This essentially eliminates the possibility of increased stress changes in the ship section that may be caused by attitude adjustment, which is beneficial to protecting the structure of the ship section. Attached Figure Description

[0064] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1Flow chart of a method for docking and attitude adjustment of a ship section according to an embodiment of the present application;

[0066] Figure 2 Schematic diagram of the deformation trend of the outer turning of both sides of a U-shaped ship section;

[0067] Figure 3 Schematic diagram of the deformation trend of the front-back torsion of a U-shaped ship section;

[0068] Figure 4 Schematic layout diagram of the pose judgment points, measurement points and feature points of the ship section of the present application;

[0069] Figure 5 Schematic layout diagram of the rigid points, measurement points and feature points of the ship section of the present application.

[0070] In the figure: 100, ship section. Detailed implementation mode

[0071] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Usually, the components of the embodiments of the present application described and shown in the accompanying drawings here can be arranged and designed in various different configurations.

[0072] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts belong to the scope of protection of the present application.

[0073] It should be noted that the technical features in all the following embodiments can be combined or interchanged with each other on the premise of not conflicting.

[0074] It should be further noted that O, M j , M, T o , P, M0, P0, O0, T p , N j , N, O n , P n , N t , O [[ID=4)8]] t , P t , ΔP, M tem , N tem etc. all represent matrices, and other characters not specifically described can be understood according to the context.

[0075] Embodiment 1:

[0076] like Figure 1 As shown, a method for docking and adjusting the attitude of ship sections is first provided, including the following steps S10 to S70.

[0077] S10. Divide the ship's main sections into rigid zones, and establish the rigid stationary point cloud O of the ship's main sections and the measurement point cloud M of each rigid zone. j The simulation yielded a point cloud M for each measurement point. j The mapping matrix T from the total cloud of measurement points M to the rigid fixed point cloud O. o A rigid fixed point cloud O contains multiple rigid points, and the relative positions of these rigid points change within a predetermined range; this can also be described as maintaining relative rigidity locally within a rigid partition. M j This represents the measurement point cloud of the j-th rigid partition.

[0078] The mapping relationship from the total cloud of measurement points M to the rigid fixed point cloud O is as follows:

[0079] O = M * T o =[M1…M j ]*T o (1-1)

[0080] It should be noted that the deformation pattern of the ship section (i.e., the deformation state that the ship section tends to undergo under the action of the lifting force of the docking vehicle roof and its own weight) can be determined through finite element analysis or historical experience data. Based on the deformation pattern, the rigid zones of the ship section can be divided. Furthermore, the relative positional changes between multiple rigid points are within a predetermined range, which can be 0.5mm, 1mm, 2mm, etc., or can be determined according to the actual required error standard. This means that the spatial distance between any two rigid points does not exceed 0.5mm, 1mm, or 2mm, etc., further indicating that during the docking process of the ship section, the entire rigid stationary point cloud changes as a whole, with essentially no internal change, and can be equated to a virtual, integral rigid structure. Of course, in the optimal solution, the relative positional changes between multiple rigid points should remain constant; however, due to limitations of the operating environment and the structure of the ship section, there will inevitably be some error.

[0081] S20. Establish the pose determination point cloud P of the ship section. This is done by inputting a set of measurement point clouds M0 and pose determination point clouds P0 obtained at the same time and in the same attitude. Then, the mapping relationship matrix T is used to determine the pose determination point cloud P. o Calculate a rigid, stationary point cloud O0, and then calculate the mapping matrix T from the rigid, stationary point cloud O0 to the pose of the point cloud P0. p .

[0082] The mapping relationship between the total cloud of measured points M0 and the rigid fixed point cloud O0 is as follows:

[0083] O0=M0*T o =[M1…M j ]*T o (1-2)

[0084] The mapping relationship between the rigid fixed point cloud O0 and the pose determination point cloud P0 is as follows:

[0085] P0 = O0 * T p (1-3)

[0086] S30. Simulate docking of the ship sections. After docking, extract multiple feature points from each rigid partition to form a feature point cloud N for each partition. j Each feature point cloud N j The total feature point cloud N is formed by mapping relationship matrix T o Obtain the rigid fixed point cloud O after docking n , from rigid fixed point cloud O n Through the mapping relationship matrix T p The pose determination point cloud P after docking is obtained n .

[0087] In step S30, N = [N1…N j Substituting this into formula (1-1), we obtain the rigid fixed point cloud O after docking. n The formula is as follows:

[0088] O n =N*T o =[N1…N j ]*T o (1-4)

[0089] O n Substituting into formula (1-3), we obtain the pose determination point cloud P after docking. n The formula is as follows:

[0090] P n =O n *T p (1-5)

[0091] S40. Before the ship section is docked, measure and acquire the real-time data of the total cloud of measurement points M. Find the measurement point in the total cloud of measurement points M that has the highest matching degree with the feature point in the total cloud of feature points N, and assign the measurement point with the highest matching degree and the feature point the same number.

[0092] S50. Perform homogeneous transformation matching from the total cloud of measurement points M to the total cloud of feature points N. The homogeneous transformation matrix is ​​T. Evaluate the reliability of measurement points with the same index as the feature points, and then determine whether the reliability of the total cloud of measurement points M meets the predetermined requirements. If the predetermined requirements are not met, repeat steps S40 and S50. If the predetermined requirements are met, continue to step S60.

[0093] S60. Multiply the total feature point cloud N by the inverse of the homogeneous transformation matrix T to obtain the real-time total feature point cloud N of the ship section under the current attitude. t The real-time feature point total cloud N t With mapping relation matrix T o Multiplication yields the real-time rigid fixed point cloud O. t , to transform real-time rigid fixed point cloud O t With mapping relation matrix T p Multiplication yields the real-time pose determination point cloud P. t Calculate the real-time pose determination point cloud P t Pole cloud P after docking n The pose deviation ΔP; the pose deviation ΔP includes the positional deviation of each real-time pose judgment point and the pose judgment point after docking in three-dimensional space.

[0094] S70. Convert the position deviation ΔP into an attitude adjustment command; send the attitude adjustment command to the docking vehicle used for docking the ship section so that the docking vehicle can make corresponding actions, or provide the attitude adjustment command to the staff for reference to adjust the attitude of the ship section.

[0095] The above scheme introduces the concept of rigid fixed points, expanding the types of ship sections that can be docked from small rigid sections to large flexible sections, no longer limited to small rigid sections. Using pose judgment points as reference data for adjustment commands, users do not need to understand the specific docking calculation principles; they only need to obtain the real-time measurement point cloud. They can then calculate the real-time parameters of the pose judgment point cloud with a finite number of pose judgment points using the aforementioned method, and compare it with the pose judgment point cloud after docking to calculate the pose deviation. Workers can then adjust the pose of the ship sections based on the pose deviation, catering to the adjustment habits and intuition of on-site construction personnel, lowering the threshold of the docking process, reducing time and manpower consumption, and improving shipbuilding efficiency.

[0096] It should be noted that the measurement points and feature points within the same rigid partition should be as far apart as possible and form irregular shapes.

[0097] Furthermore, the position and orientation determination points are physical points on the ship's main section, so that staff can conduct on-site measurements. These points include at least the bow and stern bottom deflection points and the four horizontal points at the top corners of the ship's main section.

[0098] Furthermore, the rigid partitions are symmetrically distributed along the mid-longitudinal section of the ship section, and the measurement points in the symmetrical rigid partitions are also configured to be symmetrically distributed along the mid-longitudinal section of the ship section.

[0099] Furthermore, all the fixed points of the rigid fixed point cloud are configured within the mid-longitudinal section of the ship's main section.

[0100] Further, step S40 includes the following steps:

[0101] S41. Calculate the distance feature set of each feature point relative to the other feature points;

[0102] S42. Measure the coordinate data of the measurement points of the ship section in real time, and calculate the distance feature set of each measurement point relative to the other measurement points according to the rules in step S41.

[0103] S43. Establish a similarity judgment voting mechanism between measurement points and feature points: If there are distance features in the distance feature set between a feature point and a measurement point that differ within a predetermined difference, then the similarity score between the two is increased by one; calculate the matching degree between each measurement point and each feature point according to the aforementioned rules.

[0104] S44. Assign the same number to the feature point and the measurement point with the highest matching degree.

[0105] For example, suppose the feature point cloud N corresponds to the a-th rigid partition. a Assume that the rigid partition a has k feature points. The distance between points l and j is D. l,j Then the distance feature set corresponding to point l is:

[0106]

[0107] Assume that the a-th rigid partition has q measurement points, forming a measurement point cloud M. a Assume the distance between the m-th and n-th measurement points is H. m,n Then the distance feature set corresponding to point m is:

[0108]

[0109] The degree of matching G between the measurement point m and the feature point l within the rigid partition a is determined by the distance of the feature set. ml .like Ma S m and Na S l There exist matching elements that satisfy |H m,n -D l,j |<α (α is a predetermined difference or error threshold, representing the boundary value of two distances being similar, and the unit can be mm), then G mlAdd one, the specific calculation method for the matching degree is given by the following formula:

[0110]

[0111] In the formula The term indicates that its internal values ​​are rounded down.

[0112] The matching degree matrix of size k*q can be constructed using the above method, as follows:

[0113]

[0114] The index with the highest matching score in the m-th row max G ml , indicating that the feature point index that best matches the measurement point m is l. max G ml =max(G m1 G m2 ...G mk ).

[0115] Furthermore, the feature point and measurement point with the highest matching degree are assigned the same number to facilitate subsequent steps.

[0116] Steps S41 to S44 above propose a method for pairing measurement points with feature points in the form of voting, which solves the problems of measurement point loss and disordered order during the measurement process. At the same time, the voting score can also be used as a coarse matching judgment standard for subsequent quality assessment of measurement points.

[0117] Preferably, if similar points exist within a feature point, there will be cases where the voting scores for the same measurement point are the same. Therefore, when feature points are arranged according to regular patterns such as axial symmetry and rotational symmetry, feature points at different positions need to be matched separately.

[0118] Furthermore, the first implementation of step S50 includes the following steps:

[0119] S511. In each rigid partition, select no fewer than three pairs of measurement points and feature points with the highest similarity. The measurement points selected in each rigid partition form the temporary measurement point total cloud M of the current partition. tem The feature points selected from each rigid partition form a temporary feature point cloud N. tem Calculate the total cloud M from the temporary measurement points in each partition. tem Total cloud N to temporary feature point tem Temporary homogeneous transformation matrix T j ;T j This represents the temporary homogeneous transformation matrix corresponding to the j-th rigid partition;

[0120] S512, transfer each measurement point cloud Mj Use the corresponding temporary homogeneous transformation matrix T j Perform the transformation;

[0121] S513. Set a matching error threshold between the measurement points and feature points. Match the transformed measurement points with the feature points. If the matching error is greater than the matching error threshold, exclude the corresponding measurement points and feature points to obtain the corrected measurement point cloud M for each rigid partition. a Total cloud of corrected feature points N a One way to exclude corresponding measurement points and feature points is to mark them and then ignore the marked data in subsequent calculations.

[0122] S514, Calculate the point cloud M of each rigid partition from the calibration measurement point cloud. a To correct the feature point cloud N a The corrected homogeneous transformation matrix T a ;T a This represents the correction homogeneous transformation matrix corresponding to the a-th rigid partition;

[0123] S515. Calculate the number K of measurement points greater than the matching error threshold within each rigid partition. The specific expression for K is:

[0124]

[0125] Where q represents the measured point cloud M j The number of measurement points in the middle;

[0126] ɑ represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points;

[0127] m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ;

[0128] n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ;

[0129] Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points;

[0130] The term indicates that its internal values ​​are rounded down;

[0131] S516. Determine whether the ratio of the K value in each rigid partition to the number of measurement points in the corresponding rigid partition is less than 0.3; if it is less than 0.3, then the reliability of the total cloud M of the measurement points meets the predetermined requirements, and the correction homogeneous transformation matrix T corresponding to all rigid partitions is determined. a Together they form a homogeneous transformation matrix T; if there is a case greater than 0.3, it indicates that the reliability of the total cloud M of the measurement points does not meet the predetermined requirements, and steps S40 and S50 need to be repeated.

[0132] Furthermore, the second implementation of step S50 includes the following steps:

[0133] S521, Set the measurement point cloud M in each rigid partition. j To feature point cloud N j The partition homogeneous transformation matrix is ​​T a , where a = j;

[0134] S522. Calculate the homogeneous transformation matrix of the rigid partition point cloud when the error value function of each rigid partition reaches its minimum value. a The error value function is:

[0135]

[0136] Where q represents the measured point cloud M j The number of measurement points in the middle;

[0137] α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points;

[0138] m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ;

[0139] n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ;

[0140] Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points;

[0141] The term indicates that its internal values ​​are rounded down;

[0142] S523. Calculate the number K of measurement points greater than the matching error threshold within each rigid partition. The specific expression for K is:

[0143]

[0144] Among them, T a This refers to the partition homogeneous transformation matrix calculated in step S522;

[0145] q represents the measured point cloud M j The number of measurement points in the middle;

[0146] α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points;

[0147] m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ;

[0148] n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ;

[0149] Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points;

[0150] The term indicates that its internal values ​​are rounded down;

[0151] S524. Determine whether the ratio of the K value in each rigid partition to the number of measurement points in the corresponding rigid partition is less than 0.3, and whether the minimum value of the error value function is less than or equal to a predetermined value; if they are all less than 0.3, and the error value function is less than or equal to the predetermined value, then the reliability of the total cloud M of the measurement points meets the predetermined requirements, and the partition homogeneous transformation matrix T corresponding to all rigid partitions is determined. a Together they form a homogeneous transformation matrix T; otherwise, it indicates that the reliability of the total cloud M at the measurement point does not meet the predetermined requirements, and steps S40 and S50 need to be repeated.

[0152] Preferably, the ratio of the K value in each rigid partition to the number of measurement points in the corresponding rigid partition is less than 0.2 to ensure higher reliability of the measurement points.

[0153] Step S50 of the technical solution of this application is equivalent to proposing a quantifiable docking measurement data quality assessment method. Users can quickly assess the accuracy and quality of measurement points during docking, thereby ensuring that the data obtained from the measurement points meet the requirements. Consequently, the attitude deviation and other related parameters obtained in subsequent steps can meet the actual attitude adjustment requirements, thus ensuring that the attitude adjustment parameters are appropriate. This basically eliminates the possibility of increased stress changes in the ship section that may be caused by attitude adjustment, which is beneficial to protecting the structure of the ship section.

[0154] Example 2:

[0155] This embodiment provides an application example of the ship section docking and attitude adjustment method in Embodiment 1.

[0156] first step:

[0157] Before docking the ship's main sections, the type of the main section 100 is determined, and the deformation mode of the main section under the locomotive jacking condition is clarified as fore-and-aft torsion (e.g., ...). Figure 3 (as shown) and lateral eversion (as shown) Figure 2 (As shown). The outward flaring deformation mode on both sides is effectively suppressed during actual docking due to the addition of a spacer beam spanning the entire section; therefore, it is generally not considered when dividing the rigid zones. For example... Figure 4 and Figure 5 As shown, based on the deformation pattern of the ship section 100, two symmetrically distributed rigid zones can be divided on the left and right sides of the U-shaped section. The left rigid zone is denoted as M1, and the right rigid zone is denoted as M2. The total cloud of measurement points is M = [M1 M2]. The measurement points arranged in the middle of M1 and M2 are basically symmetrical. Then, the line connecting the symmetrical measurement points of the two rigid zones is the rigid point. All rigid points constitute a rigid fixed point cloud O. It is always located within the mid-longitudinal section of the ship section, and the relative positions between the internal rigid points remain basically unchanged.

[0158] The simulation yields the mapping relationship between the total cloud of measurement points M and the rigid fixed point cloud O. The mapping relationship formula is as follows:

[0159] O = M * T o =[M1 M2]*T o (2-1)

[0160] This allows us to obtain the data from each measurement point cloud M. j The mapping matrix T from the total cloud of measurement points M to the rigid fixed point cloud O. o .

[0161] Step Two:

[0162] like Figure 4As shown, the bow and stern bottom deflection points and the four horizontal points at the top corners of the ship section 100 are selected as pose judgment points, and the pose judgment point cloud P of the ship section is established. The total measurement point cloud M0 = [M] is obtained by measuring the same pose at the same time. 01 M 02 ] and pose determination point cloud P0. Total measured point cloud M0 = [M 01 M 02 Substituting into formula (2-1), we obtain the corresponding rigid fixed point cloud O0, that is:

[0163] O0=M0*T o =[M 01 M 02 ]*T o (2-2)

[0164] Then, the mapping relationship between the rigid stationary point cloud O0 and the pose determination point cloud P0 is calculated. The mapping relationship formula is as follows:

[0165] P0 = O0 * T p (2-3)

[0166] This yields the mapping matrix T from the rigid stationary point cloud O0 to the pose of the point cloud P0. p .

[0167] Step 3:

[0168] Simulated docking can be performed using simulation software or through actual measurement and calculation. After simulated docking, multiple feature points are extracted from the two rigid partitions to form feature point clouds N1 and N2 for each rigid partition, and the total feature point cloud N = [N1 N2].

[0169] Substituting N = [N1 N2] into formula (2-1), we obtain the rigid fixed point cloud O after docking. n :

[0170] O n =N*T o =[N1 N2]*T o (2-4)

[0171] O n Substituting into formula (2-3), we obtain the pose judgment point cloud P after docking. n :

[0172] P n =O n *T p (2-5)

[0173] Step 4:

[0174] The measurement point total cloud M is obtained in real time. The measurement point with the highest matching degree with the feature point in the feature point total cloud N is found in the measurement point total cloud M, and the measurement point with the highest matching degree and the feature point are assigned the same index.

[0175] Specifically, firstly, the distance feature set of each feature point in each partition relative to all other feature points is calculated. Figure 4 and Figure 5 Taking the rigid partition on the left side as an example, let it be the first rigid partition. The feature point cloud of the first rigid partition is N1, and the l-th feature point N within it is... 1,l Distance feature set N1 S l for:

[0176]

[0177] Then calculate the distance feature set of each measurement point relative to the other measurement points. Figure 4 and Figure 5 Taking the left-hand rigid partition as an example, denoted as the first rigid partition, the measurement point cloud of the first rigid partition is M1, and the m-th measurement point M within it is... 1,m Distance feature set M1 S m for:

[0178]

[0179] Furthermore, a similarity judgment voting mechanism between measurement points and feature points is established: if there are distance features in the distance feature set between a feature point and a measurement point that differ within a predetermined difference, the similarity score between the two is increased by one; the matching degree between each measurement point and each feature point is calculated according to the aforementioned rules.

[0180] For example, the matching degree G between the first rigid partition measurement point m and the feature point l ml The calculation formula is:

[0181]

[0182] like M1 S m and N1 S l There exist elements that satisfy (α is the error threshold, representing the boundary value between two similar distances, usually in mm), then G ml Add one. G ml The physical meaning is the number of distance values ​​with similar lengths between feature point l and measurement point m in the distance data, corresponding to G. mlThe larger the value, the more similar the distance vectors between the two points, and the greater the probability that they are corresponding points, which means a higher degree of matching.

[0183] Assuming the first rigid partition has q measurement points and 7 feature points, a matching degree matrix of size 7*q can be constructed as follows:

[0184]

[0185] like Figure 4 and Figure 5 The calculation of the matching degree between the measurement points and feature points of the rigid partition shown on the right can be referenced from the above calculation.

[0186] Step 5:

[0187] Based on the matching degree information between the measurement points and feature points in the preceding steps, solve for the homogeneous transformation matrices T1 and T2 from the measurement point cloud to the feature point cloud for the first (left) and second (right) rigid partitions. T1 and T2 satisfy:

[0188]

[0189] Calculate the error value functions F1 and F2 corresponding to the homogeneous transformation matrix of each partition:

[0190]

[0191] The number of measurement points K1 and K2 with matching errors greater than the error threshold are also given:

[0192]

[0193] Determine if the ratio of the K value in each rigid partition to the number of measurement points in that rigid partition is less than 0.2, and if the minimum value of the error value function is less than a predetermined value, such as 1 mm or 2 mm. If the ratio is less than 0.2 and the error value function is less than the predetermined value, then the reliability of the total cloud M of the measurement points meets the predetermined requirements. Then, the partition homogeneous transformation matrix T corresponding to all rigid partitions is determined. a Together they form a homogeneous transformation matrix T; otherwise, it indicates that the reliability of the total cloud M at the measurement point does not meet the predetermined requirements, and steps four and five need to be repeated.

[0194] Step 6:

[0195] Multiplying the total feature point cloud N by the inverse of the homogeneous transformation matrix T yields the real-time total feature point cloud N of the ship section under its current attitude. t N t =N*T o =[N1 N2]*T o Then, the real-time feature point total cloud N tWith mapping relation matrix T o Multiplication yields the real-time rigid fixed point cloud O. t ,Right now:

[0196] O t =N t *T o =[T1N1 T2N2]*T o (2-13)

[0197] Then, the real-time rigid fixed point cloud O t With mapping relation matrix T p Multiplication yields the real-time pose determination point cloud P. t ,Right now:

[0198] P t =O t *T p =N t *T o *T p =[T1N1 T2N2]*T o *T p (2-14)

[0199] The point cloud for post-docking pose determination has been calculated before docking. n Then the current real-time pose determination point cloud P t The point cloud for pose determination after docking is P n The pose deviation ΔP is as follows:

[0200] ΔP=P n -P t =[(I-T1)N1 (I-T2)N2]*T o *T p (2-15)

[0201] The pose deviation ΔP includes the positional deviation in three-dimensional space between each real-time pose judgment point and the pose judgment point after docking. Specifically, it is represented by the deviation of the bow / stern bottom deflection point, the four horizontal points of the deck, and the points after docking in the X / Y / Z directions. These coordinate information deviation values ​​are consistent with the adjustment personnel's thinking and can also be directly verified on-site using measuring equipment such as a square and a total station.

[0202] Step 7:

[0203] The posture deviation ΔP is converted into an attitude adjustment command. This command is then sent to the docking vehicle used for docking ship sections, enabling it to perform corresponding actions. Alternatively, the command can be provided to staff for reference in adjusting the attitude of the ship sections. The attitude adjustment command may include specific rotation points, rotation angles, and displacement distances. These parameters are entered into the docking vehicle's control system to control its movements. Staff can also perform manual fine-tuning based on these parameters.

[0204] After step seven, you can manually measure the pose judgment point. If the pose deviation ΔP between the docked pose judgment point and the target pose is within the allowable standard range, the docking is complete. If it is not within the standard range, you need to repeat steps four through seven.

[0205] Example 3:

[0206] This embodiment provides a ship section docking and attitude adjustment system that can implement the methods in Embodiments 1 and 2.

[0207] A ship section docking and attitude adjustment system is characterized by comprising a point cloud establishment module, a mapping relationship calculation module, a simulated docking module, a matching module, a reliability judgment module, a deviation calculation module, and a conversion module.

[0208] The point cloud creation module is used to divide the ship section into rigid zones, and to create the rigid stationary point cloud O of the ship section and the measurement point cloud M of each rigid zone. j This method establishes a point cloud P for determining the pose of the ship section; it is also used to extract multiple feature points in each rigid partition after simulated docking of the ship section to form a feature point cloud N for each partition. j The mapping relationship calculation module is used to simulate and obtain the mapping relationship from each measured point cloud M. j The mapping matrix T from the total cloud of measurement points M to the rigid fixed point cloud O. o It is also used to input a set of total measurement point cloud M0 and pose determination point cloud P0 obtained at the same time and in the same posture, and then to use the mapping relationship matrix T o Calculate a rigid, stationary point cloud O0, and then calculate the mapping matrix T from the rigid, stationary point cloud O0 to the pose of the point cloud P0. p The simulated docking module is used to simulate the docking of ship sections, and will be generated by each feature point cloud N. j The total feature point cloud N is formed by mapping relationship matrix T o Obtain the rigid fixed point cloud O after docking n , from rigid fixed point cloud O n Through the mapping relationship matrix T p The pose determination point cloud P after docking is obtained nThe matching module is used to find the measurement point with the highest matching degree with the feature point in the feature point total cloud N in the measurement point total cloud M based on the real-time measurement point total cloud M, and assign the measurement point with the highest matching degree and the feature point the same index. The reliability judgment module is used to perform homogeneous transformation matching from the measurement point total cloud M to the feature point total cloud N, with homogeneous transformation matrix T, to evaluate the reliability of the measurement point with the same index as the feature point, and then determine whether the reliability of the measurement point total cloud M meets the predetermined requirements. The deviation calculation module is used to multiply the feature point total cloud N by the inverse matrix of the homogeneous transformation matrix T to obtain the real-time feature point total cloud N under the current attitude of the ship section. t The real-time feature point total cloud N t With mapping relation matrix T o Multiplication yields the real-time rigid fixed point cloud O. t , to transform real-time rigid fixed point cloud O t With mapping relation matrix T p Multiplication yields the real-time pose determination point cloud P. t Calculate the real-time pose determination point cloud P t Pole cloud P after docking n The pose deviation ΔP is converted into a pose adjustment command by the conversion module.

[0209] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the methods in Embodiments 1 and 2.

[0210] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the methods in Embodiments 1 and 2.

[0211] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for docking and adjusting the attitude of ship sections, characterized in that, Includes the following steps: S10. Divide the ship's main sections into rigid zones, and establish the rigid stationary point cloud O of the ship's main sections and the measurement point cloud M of each rigid zone. j The simulation yielded a point cloud M for each measurement point. j The mapping matrix T from the total cloud of measurement points M to the rigid fixed point cloud O. o The rigid fixed point cloud O contains multiple rigid points, and the relative positions of these rigid points vary within a predetermined range; M j Represents the measurement point cloud of the j-th rigid partition; S20. Establish the pose determination point cloud P of the ship section. This is done by inputting a set of measurement point clouds M0 and pose determination point clouds P0 obtained at the same time and in the same attitude. Then, the mapping relationship matrix T is used to determine the pose determination point cloud P. o Calculate a rigid, stationary point cloud O0, and then calculate the mapping matrix T from the rigid, stationary point cloud O0 to the pose of the point cloud P0. p ; S30. Simulate docking of the ship sections. After docking, extract multiple feature points from each rigid partition to form a feature point cloud N for each partition. j Each feature point cloud N j The total feature point cloud N is formed by mapping relationship matrix T o Obtain the rigid fixed point cloud O after docking n , from rigid fixed point cloud O n Through the mapping relationship matrix T p The pose determination point cloud P after docking is obtained n ; S40. Before the docking of the ship section is completed, based on the real-time measurement point cloud M, find the measurement point with the highest matching degree with the feature point in the feature point cloud N in the measurement point cloud M, and assign the measurement point with the highest matching degree and the feature point the same number. S50. Perform homogeneous transformation matching from the total cloud of measurement points M to the total cloud of feature points N. The homogeneous transformation matrix is ​​T. Evaluate the reliability of measurement points with the same index as the feature points, and then determine whether the reliability of the total cloud of measurement points M meets the predetermined requirements. If the predetermined requirements are not met, repeat steps S40 and S50. If the predetermined requirements are met, continue to step S60. S60. Multiply the total feature point cloud N by the inverse of the homogeneous transformation matrix T to obtain the real-time total feature point cloud N of the ship section under the current attitude. t The real-time feature point total cloud N t With mapping relation matrix T o Multiplication yields the real-time rigid fixed point cloud O. t , to transform real-time rigid fixed point cloud O t With mapping relation matrix T p Multiplication yields the real-time pose determination point cloud P. t Calculate the real-time pose determination point cloud P t Pole cloud P after docking n The pose deviation ΔP; the pose deviation includes the positional deviation of each real-time pose judgment point and the pose judgment point after docking in three-dimensional space. S70. Convert the positional deviation ΔP into an attitude adjustment command; send the attitude adjustment command to the docking vehicle used for docking the ship section so that the docking vehicle can make corresponding actions, or provide the attitude adjustment command to the staff for reference to adjust the attitude of the ship section. Step S40 includes the following steps: S41. Calculate the distance feature set of each feature point relative to the other feature points; S42. Measure the coordinate data of the measurement points of the ship section in real time, and calculate the distance feature set of each measurement point relative to the other measurement points according to the rules in step S41. S43. Establish a similarity judgment voting mechanism between measurement points and feature points: If there are distance features in the distance feature set between a feature point and a measurement point that differ within a predetermined difference, then the similarity score between the two is increased by one; calculate the matching degree between each measurement point and each feature point according to the aforementioned rules. S44. Assign the same number to the feature point and the measurement point with the highest matching degree; Step S50 includes the following steps: S511. In each rigid partition, select no fewer than three pairs of measurement points and feature points with the highest similarity. The measurement points selected in each rigid partition form the temporary measurement point total cloud M of the current partition. tem The feature points selected from each rigid partition form a temporary feature point cloud N. tem Calculate the total cloud M from the temporary measurement points in each partition. tem Total cloud N to temporary feature point tem Temporary homogeneous transformation matrix T j ;T j This represents the temporary homogeneous transformation matrix corresponding to the j-th rigid partition; S512, Transfer each measurement point cloud M j Use the corresponding temporary homogeneous transformation matrix T j Perform the transformation; S513. Set a matching error threshold between the measurement points and feature points. Match the transformed measurement points with the feature points. If the matching error is greater than the matching error threshold, exclude the corresponding measurement points and feature points to obtain the corrected measurement point cloud M for each rigid partition. a and correction feature point total cloud N a ; S514. Calculate the point cloud M of each rigid partition from the calibration measurement point cloud. a To correct the feature point cloud N a The corrected homogeneous transformation matrix T a ;T a This represents the correction homogeneous transformation matrix corresponding to the a-th rigid partition; S515. Calculate the number K of measurement points greater than the matching error threshold within each rigid partition. The specific expression for K is: Where q represents the measured point cloud M j The number of measurement points in the middle; α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points; m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ; n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ; Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points; The term indicates that its internal values ​​are rounded down; S516. Determine whether the ratio of the K value in each rigid partition to the number of measurement points in the corresponding rigid partition is less than 0.3; if it is less than 0.3, then the reliability of the total cloud M of the measurement points meets the predetermined requirements, and the correction homogeneous transformation matrix T corresponding to all rigid partitions is determined. a Together they form a homogeneous transformation matrix T; if there is a case greater than 0.3, it indicates that the reliability of the total cloud M of the measurement points does not meet the predetermined requirements, and steps S40 and S50 need to be repeated.

2. The method for docking and adjusting the attitude of ship sections according to claim 1, characterized in that, The pose determination points are physical points on the ship's main body, including at least the bow and stern bottom deflection points and the four horizontal points at the top corners of the ship's main body.

3. The method for docking and adjusting the attitude of ship sections according to claim 1, characterized in that, The rigid partitions are symmetrically distributed along the mid-longitudinal section of the ship section, and the measurement points in the symmetrical rigid partitions are also configured to be symmetrically distributed along the mid-longitudinal section of the ship section.

4. The method for docking and adjusting the attitude of ship sections according to claim 3, characterized in that, All the fixed points of the rigid fixed point cloud are configured within the mid-longitudinal section of the ship's main section.

5. A ship section docking and attitude adjustment system, characterized in that, include: The point cloud creation module is used to divide the ship section into rigid zones, and to create the rigid stationary point cloud O of the ship section and the measurement point cloud M of each rigid zone. j This method establishes a point cloud P for determining the pose of the ship section; it is also used to extract multiple feature points in each rigid partition after simulated docking of the ship section to form a feature point cloud N for each partition. j ; The mapping relationship calculation module is used to simulate and obtain the mapping relationship from each measured point cloud M. j The mapping matrix T from the total cloud of measurement points M to the rigid fixed point cloud O. o It is also used to input a set of total measurement point cloud M0 and pose determination point cloud P0 obtained at the same time and in the same posture, and then to use the mapping relationship matrix T o Calculate a rigid, stationary point cloud O0, and then calculate the mapping matrix T from the rigid, stationary point cloud O0 to the pose of the point cloud P0. p ; The simulated docking module is used to simulate the docking of ship sections, and will be generated by each feature point cloud N. j The total feature point cloud N is formed by mapping relationship matrix T o Obtain the rigid fixed point cloud O after docking n , from rigid fixed point cloud O n Through the mapping relationship matrix T p The pose determination point cloud P after docking is obtained n ; The matching module is used to find the measurement point with the highest matching degree with the feature point in the total feature point cloud N in the total measurement point cloud M based on the real-time data of the total measurement point cloud M, and assign the measurement point with the highest matching degree and the feature point the same number. The steps include: S41. Calculate the distance feature set of each feature point relative to the other feature points; S42. Measure the coordinate data of the measurement points of the ship section in real time, and calculate the distance feature set of each measurement point relative to the other measurement points according to the rules in step S41. S43. Establish a similarity judgment voting mechanism between measurement points and feature points: If there are distance features in the distance feature set between a feature point and a measurement point that differ within a predetermined difference, then the similarity score between the two is increased by one; calculate the matching degree between each measurement point and each feature point according to the aforementioned rules. S44. Assign the same number to the feature point and the measurement point with the highest matching degree; The reliability assessment module performs homogeneous transformation matching from the total cloud of measurement points M to the total cloud of feature points N, with a homogeneous transformation matrix of T. It evaluates the reliability of measurement points with the same index as the feature points, thereby determining whether the reliability of the total cloud of measurement points M meets predetermined requirements. Its steps include:

511. In each rigid partition, select no fewer than three pairs of measurement points and feature points with the highest similarity. The measurement points selected in each rigid partition form the temporary measurement point total cloud M for the current partition. tem The feature points selected from each rigid partition form a temporary feature point cloud N. tem Calculate the total cloud M from the temporary measurement points in each partition. tem Total cloud N to temporary feature point tem Temporary homogeneous transformation matrix T j ;T j This represents the temporary homogeneous transformation matrix corresponding to the j-th rigid partition; S512, Transfer each measurement point cloud M j Use the corresponding temporary homogeneous transformation matrix T j Perform the transformation; S513. Set a matching error threshold between the measurement points and feature points. Match the transformed measurement points with the feature points. If the matching error is greater than the matching error threshold, exclude the corresponding measurement points and feature points to obtain the corrected measurement point cloud M for each rigid partition. a and correction feature point total cloud N a ; S514. Calculate the point cloud M of each rigid partition from the calibration measurement point cloud. a To correct the feature point cloud N a The corrected homogeneous transformation matrix T a ;T a This represents the correction homogeneous transformation matrix corresponding to the a-th rigid partition; S515. Calculate the number K of measurement points greater than the matching error threshold within each rigid partition. The specific expression for K is: Where q represents the measured point cloud M j The number of measurement points in the middle; α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points; m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ; n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ; Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points; The term indicates that its internal values ​​are rounded down; S516. Determine whether the ratio of the K value in each rigid partition to the number of measurement points in the corresponding rigid partition is less than 0.3; if it is less than 0.3, then the reliability of the total cloud M of the measurement points meets the predetermined requirements, and the correction homogeneous transformation matrix T corresponding to all rigid partitions is determined. a Together they form a homogeneous transformation matrix T; if there is a case greater than 0.3, it indicates that the reliability of the total cloud M of the measurement points does not meet the predetermined requirements, and the steps executed by the matching module and the reliability judgment module need to be repeated. The deviation calculation module is used to multiply the total feature point cloud N by the inverse of the homogeneous transformation matrix T to obtain the real-time total feature point cloud N under the current attitude of the ship section. t The real-time feature point total cloud N t With mapping relation matrix T o Multiplication yields the real-time rigid fixed point cloud O. t , to transform real-time rigid fixed point cloud O t With mapping relation matrix T p Multiplication yields the real-time pose determination point cloud P. t Calculate the real-time pose determination point cloud P t Pole cloud P after docking n The pose deviation ΔP; The conversion module is used to convert the pose deviation ΔP into a pose adjustment command.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

8. A method for docking and adjusting the attitude of ship sections, characterized in that, Includes the following steps: S10. Divide the ship's main sections into rigid zones, and establish the rigid stationary point cloud O of the ship's main sections and the measurement point cloud M of each rigid zone. j The simulation yielded a point cloud M for each measurement point. j The mapping matrix T from the total cloud of measurement points M to the rigid fixed point cloud O. o The rigid fixed point cloud O contains multiple rigid points, and the relative positions of these rigid points vary within a predetermined range; M j Represents the measurement point cloud of the j-th rigid partition; S20. Establish the pose determination point cloud P of the ship section. This is done by inputting a set of measurement point clouds M0 and pose determination point clouds P0 obtained at the same time and in the same attitude. Then, the mapping relationship matrix T is used to determine the pose determination point cloud P. o Calculate a rigid, stationary point cloud O0, and then calculate the mapping matrix T from the rigid, stationary point cloud O0 to the pose of the point cloud P0. p ; S30. Simulate docking of the ship sections. After docking, extract multiple feature points from each rigid partition to form a feature point cloud N for each partition. j Each feature point cloud N j The total feature point cloud N is formed by mapping relationship matrix T o Obtain the rigid fixed point cloud O after docking n , from rigid fixed point cloud O n Through the mapping relationship matrix T p The pose determination point cloud P after docking is obtained n ; S40. Before the docking of the ship section is completed, based on the real-time measurement point cloud M, find the measurement point with the highest matching degree with the feature point in the feature point cloud N in the measurement point cloud M, and assign the measurement point with the highest matching degree and the feature point the same number. S50. Perform homogeneous transformation matching from the total cloud of measurement points M to the total cloud of feature points N. The homogeneous transformation matrix is ​​T. Evaluate the reliability of measurement points with the same index as the feature points, and then determine whether the reliability of the total cloud of measurement points M meets the predetermined requirements. If the predetermined requirements are not met, repeat steps S40 and S50. If the predetermined requirements are met, continue to step S60. S60. Multiply the total feature point cloud N by the inverse of the homogeneous transformation matrix T to obtain the real-time total feature point cloud N of the ship section under the current attitude. t The real-time feature point total cloud N t With mapping relation matrix T o Multiplication yields the real-time rigid fixed point cloud O. t , to transform real-time rigid fixed point cloud O t With mapping relation matrix T p Multiplication yields the real-time pose determination point cloud P. t Calculate the real-time pose determination point cloud P t Pole cloud P after docking n The pose deviation ΔP; the pose deviation includes the positional deviation of each real-time pose judgment point and the pose judgment point after docking in three-dimensional space. S70. Convert the positional deviation ΔP into an attitude adjustment command; send the attitude adjustment command to the docking vehicle used for docking the ship section so that the docking vehicle can make corresponding actions, or provide the attitude adjustment command to the staff for reference to adjust the attitude of the ship section. Step S40 includes the following steps: S41. Calculate the distance feature set of each feature point relative to the other feature points; S42. Measure the coordinate data of the measurement points of the ship section in real time, and calculate the distance feature set of each measurement point relative to the other measurement points according to the rules in step S41. S43. Establish a similarity judgment voting mechanism between measurement points and feature points: If there are distance features in the distance feature set between a feature point and a measurement point that differ within a predetermined difference, then the similarity score between the two is increased by one; calculate the matching degree between each measurement point and each feature point according to the aforementioned rules. S44. Assign the same number to the feature point and the measurement point with the highest matching degree; Step S50 includes the following steps: S521, Set the measurement point cloud M in each rigid partition. j To feature point cloud N j The partition homogeneous transformation matrix is ​​T a , where a = j; S522. Calculate the homogeneous transformation matrix of the rigid partition point cloud when the error value function of each rigid partition reaches its minimum value. a The error value function is: Where q represents the measured point cloud M j The number of measurement points in the middle; α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points; m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ; n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ; Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points; The term indicates that its internal values ​​are rounded down; S523. Calculate the number K of measurement points greater than the matching error threshold within each rigid partition. The specific expression for K is: Among them, T a This refers to the partition homogeneous transformation matrix calculated in step S522; q represents the measured point cloud M j The number of measurement points in the middle; α represents the matching error threshold between the measurement point and the feature point, which characterizes the boundary value of similar distance between two points; m l To calibrate the measured point cloud M c The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point m. l ; n l For the partitioned feature point cloud N a The point in the middle, This indicates the direction from the origin of the ship's coordinate system to point n. l ; Measurement point m l To calibrate the measured point cloud M c N and the feature point cloud of the region a Middle feature point n l Matching points; The term indicates that its internal values ​​are rounded down; S524. Determine whether the ratio of the K value in each rigid partition to the number of measurement points in the corresponding rigid partition is less than 0.3, and whether the minimum value of the error value function is less than or equal to a predetermined value; if they are all less than 0.3 and the minimum value of the error value function is less than or equal to the predetermined value, then the reliability of the total cloud M of the measurement points meets the predetermined requirements, and the partition homogeneous transformation matrix T corresponding to all rigid partitions is determined. a Together they form a homogeneous transformation matrix T; otherwise, it indicates that the reliability of the total cloud M at the measurement point does not meet the predetermined requirements, and steps S40 and S50 need to be repeated.

Images