A method and system for precision control of a modular structure of a ship

Abstract

This invention relates to the field of ship structure assembly technology and discloses a method and system for precision control of ship modular structures. The method includes: generating a dynamic precision field to characterize the deviation between the actual and desired pose of the ship structure; aiming at optimal matching degree of the assembly interface, collaboratively optimizing the assembly sequence, pose adjustment path, and interface compensation amount of the ship structure based on the dynamic precision field to generate an optimal assembly strategy instruction set to guide the assembly operation of the ship structure; and converting the optimal assembly strategy instruction set into control instructions for the actuators to drive the actuators to dynamically adjust the pose of the ship structure. This invention quantifies the global deviation in real time through the dynamic precision field and uses this as a basis for multi-objective collaborative optimization of the assembly sequence, adjustment path, and compensation amount to generate globally optimal assembly instructions. Finally, iterative convergence of precision is achieved through closed-loop execution and feedback, thereby significantly improving the assembly accuracy.

Description

Technical Field

[0001] This invention relates to the field of ship structure assembly technology, and more specifically, to a method and system for precision control of ship modular structures. Background Technology

[0002] Modular ship structure is a structure that breaks down a ship into several standardized and universal independent modules according to its functions and structural attributes. Each module can complete structural manufacturing, pipeline pre-installation, equipment integration and other processes in the factory in advance, and then be transported to the slipway or dock for overall assembly. Its core is to improve construction efficiency and reduce on-site operation difficulty through modular prefabrication and integrated assembly.

[0003] Precision control during assembly is crucial because ships, as large and complex structural components, have interfaces between modules that directly determine the feasibility of closure, structural strength, and sealing performance. Substandard precision can lead to interface misalignment, welding stress concentration, and negatively impact core performance characteristics such as navigational stability and corrosion resistance. If precision cannot be optimized during assembly by incorporating compensation amounts and assembly sequences, it becomes difficult to balance the mutual influence of different modules, resulting in accumulated interface deviations. These deviations cannot be eliminated through precise compensation and must be corrected on-site through forced repairs or grinding. This not only reduces the first-pass yield and structural reliability of ship assembly but also fails to meet the high-efficiency and high-precision requirements of modular construction for large ships.

[0004] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0005] To address the problems in related technologies, this invention proposes a method and system for precision control of modular ship structures, in order to overcome the aforementioned technical problems existing in the prior art.

[0006] Therefore, the specific technical solution adopted by the present invention is as follows: According to a first aspect of the present invention, a method for precision control of a ship's modular structure is provided, the method comprising: S1. Real-time acquisition of multi-source data of the ship structure during the pre-assembly stage, and spatiotemporal alignment and fusion processing of the multi-source data to generate a dynamic accuracy field to characterize the deviation between the actual pose and the desired pose of the ship structure. S2. With the goal of achieving the optimal matching degree of the main section closure interface, the assembly sequence, posture adjustment path and interface compensation amount of the ship structure are collaboratively optimized based on the dynamic precision field to generate the optimal assembly strategy instruction set to guide the assembly operation of the ship structure. S3. Convert the optimal assembly strategy instruction set into control instructions for the actuator, drive the actuator to dynamically adjust the ship's structural attitude, and simultaneously collect the actual measurement data after dynamic adjustment and feed it back to the data fusion processing in step S1 to update the dynamic accuracy field.

[0007] According to a second aspect of the present invention, a precision control system for a modular ship structure is provided, the system comprising: The precision field construction module is used to collect multi-source data of the ship structure in real time during the pre-assembly stage, and to perform spatiotemporal alignment and fusion processing on the multi-source data to generate a dynamic precision field that characterizes the deviation between the actual pose and the expected pose of the ship structure. The assembly strategy optimization module is used to optimize the assembly sequence, posture adjustment path and interface compensation of the ship structure based on the dynamic precision field with the goal of achieving the best matching degree of the joint interface. It generates the optimal assembly strategy instruction set to guide the assembly operation of the ship structure. The assembly execution module is used to convert the optimal assembly strategy instruction set into control instructions for the actuator, drive the actuator to dynamically adjust the ship's structural attitude, and collect the actual measurement data after dynamic adjustment and feed it back to the data fusion processing process in the accuracy field construction module to update the dynamic accuracy field.

[0008] According to a third aspect of the present invention, a computer device is provided, the computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described method.

[0009] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0010] The beneficial effects of this invention are as follows: 1. This invention quantifies global deviations in real time through a dynamic precision field, and based on this, performs multi-objective collaborative optimization of assembly sequence, adjustment path and compensation amount to generate globally optimal assembly instructions. Finally, it achieves iterative convergence of precision through closed-loop execution and feedback, thereby significantly improving assembly precision, greatly reducing reliance on manual experience, reducing the time and economic cost of on-site adjustments and rework, and realizing high-precision, adaptive and intelligent control of modular ship construction.

[0011] 2. This invention utilizes spatiotemporal alignment, iterative registration, and weighted fusion technologies to ensure the global consistency and robustness of deviation perception. It also achieves the global optimal decision-making of assembly elements through collaborative optimization and finally realizes the autonomous iterative convergence of accuracy through closed-loop execution and feedback mechanisms. This significantly improves the assembly accuracy, first-time success rate, and intelligence level of modular construction of large ships, effectively reducing the reliance on human experience and the time and economic costs caused by on-site adjustments and rework.

[0012] 3. This invention combines coupled topological sorting and potential field-guided collaborative optimization strategies to perform multi-objective collaborative optimization of assembly sequence, pose path and interface compensation amount, generating a globally optimal assembly strategy instruction set. This breaks through the limitations of traditional sequential decision-making. Furthermore, through closed-loop execution and real-time feedback driven by instructions, it realizes adaptive adjustment of pose and compensation amount and iterative convergence of accuracy during the assembly process, thereby improving the success rate and accuracy consistency of large-size structure assembly. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart of a method for precision control of a modular ship structure according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a precision control system for a modular ship structure according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a computer device.

[0015] In the picture: 1. Precision field construction module; 2. Assembly strategy optimization module; 3. Assembly execution module. Detailed Implementation

[0016] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention.

[0017] According to an embodiment of the present invention, a method and system for precision control of modular ship structures are provided.

[0018] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, the ship modular structure precision control method according to an embodiment of the present invention includes: S1. Real-time acquisition of multi-source data of the ship structure during the pre-assembly stage, and spatiotemporal alignment and fusion processing of the multi-source data to generate a dynamic accuracy field that characterizes the deviation between the actual pose and the desired pose of the ship structure.

[0019] In one embodiment, the real-time acquisition of multi-source data of the ship structure during the pre-assembly stage, and the spatiotemporal alignment and fusion processing of the multi-source data to generate a dynamic accuracy field characterizing the deviation between the actual and desired poses of the ship structure, includes: S11. Integrate multi-source data of the ship structure during the pre-assembly stage collected by sensing devices, and perform timestamp synchronization and unified coordinate system processing on the multi-source fused data to generate an original pose observation sequence with consistent spatiotemporal reference.

[0020] It should be noted that the multi-source data includes high-precision discrete point 3D coordinates obtained through laser trackers and total stations, dense surface point clouds obtained through industrial photogrammetry systems, local deformation data collected by strain sensors, and data from environmental temperature and humidity sensors. This step includes: assigning a unified timestamp to all sensor data streams for synchronization; uniformly transforming the observations from various sensors based on their own coordinate systems to the global coordinate system of ship construction using a pre-calibrated transformation matrix; and finally outputting a structured pose observation sequence containing point coordinates, point clouds, and auxiliary information, consistent with the temporal and spatial reference. Spatiotemporal alignment eliminates data heterogeneity, providing standardized input for subsequent fusion processing.

[0021] S12. The original pose observation sequence is associated and matched with the preset desired pose to obtain the position deviation features of each feature point. The position deviation features of each feature point are fused to generate the fused pose estimate of the key feature points on the surface of the ship structure.

[0022] In one embodiment, the step of associating and matching the original pose observation sequence with a preset desired pose to obtain the position deviation features of each feature point, and fusing the position deviation features of each feature point to generate a fused pose estimate of key feature points on the ship structure surface includes: S121. Extract the observation coordinates of each observation feature point from the original pose observation sequence, and preliminarily correlate them with the corresponding theoretical feature point coordinates in the preset desired pose. S122. Based on the initial correlation features between feature points, the optimal spatial transformation parameters are obtained by registering the observed feature point set and the theoretical feature point set using the iterative nearest point algorithm.

[0023] In one embodiment, the step of registering the observed feature point set and the theoretical feature point set using the iterative nearest point algorithm to obtain the optimal spatial transformation parameters includes: S1221. Search for the nearest neighbor of each observed feature point in the theoretical feature point set to establish an initial corresponding point pair, and filter the initial corresponding point pairs in combination with the observation uncertainty to form a candidate corresponding point pair set. S1222. Based on the reciprocal of the observation uncertainty of each point pair in the candidate corresponding point pair set as the weight, construct and solve the weighted least squares problem, and calculate the optimal spatial transformation parameters that minimize the sum of squared weighted distances between all corresponding point pairs.

[0024] It should be noted that the observation uncertainty of each candidate point pair is quantified into a variance value, and its reciprocal is used as the weight of the point pair in the least squares problem. A weighted distance sum of squares function with rotation matrix and translation vector as optimization variables is constructed as the objective function. By solving this weighted least squares problem, such as using analytical methods like SVD or numerical iteration, the optimal spatial transformation parameters that prioritize the alignment of high-confidence point pairs are calculated. Thus, in real data with observation noise and outliers, a more robust and accurate registration result than standard least squares is obtained.

[0025] S1223. Update the pose of the observed feature point set using the optimal spatial transformation parameters, and calculate the average matching error between the updated observed feature point set and the theoretical feature point set. S1224. If the average matching error is less than the preset threshold or the maximum number of iterations is met, the final optimal spatial transformation parameters are output. Otherwise, the pose of the updated observation feature point set is used as input and the process returns to step S1221 for the next iteration calculation until the average matching error is less than the preset threshold or the maximum number of iterations is reached.

[0026] It should be noted that by assigning a greater weight to the high-precision observation points using the reciprocal of the uncertainty, the optimal spatial transformation parameters can be prioritized to align with reliable data, effectively suppressing measurement noise and disturbances caused by outliers, and significantly improving the robustness and accuracy of the registration. Secondly, through closed-loop iteration between corresponding point search, weighted parameter solution, pose update, and error judgment, adaptive optimization and refinement of the spatial transformation parameters are achieved, ensuring that the registration results converge to a stable and reliable optimal solution. This high-precision registration result lays a solid foundation for subsequent accurate calculation of pose deviations and construction of a reliable dynamic accuracy field, thereby ensuring the measurement accuracy and decision reliability of the entire ship modular structure precision control system from the source.

[0027] S123. Calculate the pose deviation of each observed feature point according to the optimal spatial transformation parameters, assign weights to each pose deviation based on preset rules and perform weighted fusion to obtain the fused pose estimate of the key feature points on the ship structure surface.

[0028] S13. Based on the fusion pose estimation of key feature points on the ship structure surface, the pose deviation of the entire ship structure is interpolated to generate a dynamic accuracy field containing spatial distribution deviation estimation.

[0029] S2. With the goal of achieving the optimal matching degree of the assembly interface of the main section, the assembly sequence, posture adjustment path and interface compensation amount of the ship structure are collaboratively optimized based on the dynamic precision field to generate the optimal assembly strategy instruction set to guide the assembly operation of the ship structure.

[0030] In one embodiment, the step of optimizing the assembly sequence, pose adjustment path, and interface compensation amount of the ship structure based on a dynamic precision field, with the goal of achieving optimal matching degree of the assembly interface, and generating an optimal assembly strategy instruction set to guide the assembly operation of the ship structure includes: S21. With the goal of achieving the optimal matching degree of the overall segment closure interface, a multi-objective optimization model is constructed based on the dynamic precision field. S22. The multi-objective optimization model is solved by combining coupled topological sorting and potential field guidance to output the optimal assembly scheme that includes the assembly sequence of the ship structure, the pose adjustment path and the interface compensation amount.

[0031] It should be noted that step S2 comprises two key steps. Step S21 involves constructing a multi-objective optimization model, aiming to transform the engineering problem into a mathematical problem with multiple mutually constraining objectives. Step S22 involves applying a collaborative optimization strategy to solve this complex mathematical problem and obtain an executable optimal solution. The fundamental purpose of this implementation is to achieve global collaborative optimization of three key decision variables: assembly sequence, pose adjustment path, and interface compensation amount, rather than traditional sequential or independent decision-making. Through collaborative optimization, the system can automatically find the optimal balance between global accuracy, adjustment efficiency, and resource cost, thereby generating a theoretically globally optimal assembly strategy instruction set to directly guide on-site operations.

[0032] Step S21 uses the dynamic precision field as the core input to construct a multi-objective optimization model that defines the optimization objective and constraints. This model is the mathematical basis and the only object for all calculations in step S22. Step S22 then solves this model by combining coupled topological sorting and potential field-guided collaborative optimization strategies. The optimal assembly scheme output is the optimal solution to the mathematical problem defined in step S21.

[0033] In one embodiment, the combined optimization strategy of coupled topological sorting and potential field guidance is used to solve the multi-objective optimization model, and the output of the optimal assembly scheme, which includes the assembly sequence of the ship structure, the pose adjustment path, and the interface compensation amount, includes: S221. Based on the geometric and technological constraints of the ship structure, construct the assembly constraint topology with the dependency relationship of assembly sequence, and simultaneously generate the pose energy field to guide pose adjustment. S222. Generate candidate assembly sequences within the framework of assembly constraint topology. Utilize the pose energy field to plan the gradient descent path and interface compensation amount for pose adjustment at each assembly step in the candidate assembly sequence, thus forming the current assembly strategy.

[0034] It should be noted that this step automatically generates a set of matching, executable, and fine-grained action plans for each possible assembly sequence. Specifically, generating candidate assembly sequences aims to explore different assembly logic sequences to find the optimal assembly process; while using the potential energy field to plan paths and compensation amounts calculates how to move the structure in the least cost and most natural way (descent along the potential energy field gradient) for each step under the selected sequence, and accurately quantifies the material that needs to be cut or filled, thereby tightly coupling the macroscopic assembly sequence decision with the microscopic execution of each step and process parameters.

[0035] This invention overcomes the limitations of traditional methods that disconnect between sequential planning and on-site adjustments by integrating assembly sequence decision-making, pose path planning, and compensation calculation in real time. This method automatically assesses the adjustment costs and achievable accuracy under different assembly sequences, proactively avoiding assembly sequences that might lead to huge adjustments or uncompensable results during the virtual planning stage, and simultaneously outputs the optimal movement command and precise process compensation parameters for each step. This not only significantly improves the overall optimization level and feasibility of the assembly strategy but also transforms the later, experience-dependent, and uncertain on-site repair work into precise, predictable, and quantifiable commands from the early stages, thereby greatly improving the first-time assembly success rate, process determinism, and overall efficiency.

[0036] In one embodiment, the step of planning the gradient descent path and interface compensation amount for pose adjustment for each assembly step in the candidate assembly sequence using the pose energy field includes: For each assembly step in the candidate assembly sequence, the current pose of the ship structure is set as the starting point of the path. Guided by the gradient descent direction of the pose energy field, several gradient descent paths for pose adjustment to reach the target assembly area are planned, and the pose at the end of the path is recorded. Calculate the residual between the end-path pose and the preset closing pose, decompose the residual into the part eliminated by the actuator and the part that needs to be compensated for by materials, and output the interface compensation amount required for the current assembly step.

[0037] S223. Evaluate the quality of the current assembly strategy using the total adjustment cost and closing accuracy as indicators, and iteratively update the assembly sequence and its corresponding pose adjustment optimization path and interface compensation amount based on the evaluation results. S224. When the iteration meets the convergence condition, output the assembly sequence, pose adjustment path and interface compensation amount that make the overall segment closing interface matching degree optimal, as the optimal assembly scheme obtained by solving.

[0038] S23. Transform the optimal assembly scheme into action instructions and process parameters that can drive the actuator, forming an optimal assembly strategy instruction set to guide the assembly operation of the ship structure.

[0039] It should be noted that this invention combines coupled topological sorting and potential field-guided collaborative optimization strategies to solve multi-objective optimization models. Its core logic lies in decomposing the complex global assembly optimization problem into two strongly coupled sub-problems: sequence decision-making and pose-compensation planning. An iterative feedback mechanism is used to achieve collaborative optimization between these two sub-problems. The fundamental purpose of this strategy is to solve the coordination problem among three mutually influential variables—assembly sequence, pose adjustment path, and interface compensation amount—which are difficult to optimize independently. Thus, under the premise of satisfying process feasibility, it automatically searches for the globally optimal or near-optimal assembly scheme that minimizes the total adjustment cost and maximizes the final joining accuracy.

[0040] The architecture of the multi-objective optimization model consists of an objective function and constraints. The objective function aims to maximize the matching degree of the merging interface of the main sections and minimize the total adjustment cost. The constraints are mainly encapsulated in two parts: one is the assembly constraint topology, which is determined by the ship's structural geometry and process logic and is used to generate feasible assembly sequences, defining the sequence network of assembly operations; the other is the potential energy field derived from the dynamic precision field and used to guide efficient pose adjustment. This potential energy field uses the high-precision region as the low potential energy valley, naturally defining the gradient descent direction from the current pose to the target pose.

[0041] The implementation principle of the multi-objective optimization model is a hierarchical-cooperative optimization mechanism: the topology sorting layer generates or searches for candidate assembly sequences within the constrained topology; for each candidate sequence, the potential energy field guiding layer quickly plans a locally optimal pose adjustment gradient descent path for each step under that sequence, and calculates the necessary interface compensation based on the residual between the pose at the end of the path and the ideal pose, thus forming a complete assembly strategy; the quality of the strategy is quantified by the total adjustment cost and the predicted merging accuracy; by iteratively updating the candidate sequences and repeating the above process, the entire system is driven to evolve towards a better strategy. The parameters and processing methods selected by this strategy are highly targeted: its core parameters are the construction rules of the potential energy field and the convergence criteria of the iterative optimization.

[0042] The discrete combinatorial optimization problem of assembly sequence is efficiently coupled with the continuous spatial optimization problem of path and compensation amount through potential energy field and iterative evaluation. The specific sub-steps of steps S221 to S224 are detailed expansions of the solver or execution engine of the multi-objective optimization model. The assembly constraint topology and potential energy field constructed in step S221 are essentially concretizations of the abstract feasibility constraints and efficiency guides in the model into computable and operable data structures. The iterative process of steps S222 to S224 is the specific algorithm process used to solve the model.

[0043] The following detailed explanation of step S2 is provided in conjunction with specific embodiments: This embodiment takes the assembly of the superstructure section of a 100,000-ton bulk carrier as an example. The section has dimensions of 28m in length, 12m in width, and 18m in height. Step S2 is executed to generate the optimal assembly strategy instruction set. The core objective is to achieve the optimal matching degree of the section assembly interface, while taking into account both adjustment efficiency and cost.

[0044] A multi-objective optimization model was constructed using a dynamic precision field as the core input. Initial precision data of each interface of the superstructure section was obtained through laser scanning, with the deviation range controlled within ±0.8mm. Combined with the real-time feedback of the assembly process precision changes from the dynamic precision field, the optimization objectives were defined as: interface matching degree of the section closure ≥98.5% and total adjustment cost ≤8h. The constraints were encapsulated as assembly constraint topology and positional energy field. The assembly constraint topology clearly defines the requirements that deck components are assembled before bulkheads at the geometric level and that welded parts must have their positions fixed at the process level. The low potential energy valleys of the positional energy field correspond to the regions where the interface precision deviation is ≤0.3mm, thus completing the construction of the multi-objective optimization model.

[0045] Based on the geometric features and process requirements of the superstructure, and combined with the load-bearing relationship of beams and columns and the avoidance of pre-installed pipelines, an assembly constraint topology is constructed to clarify the sequential dependency relationship of 23 core components; a potential energy field is generated synchronously, and the potential energy coefficient is set to be positively correlated with the accuracy deviation to guide the position adjustment to converge toward the high-precision region.

[0046] Three candidate assembly sequences are generated under the constrained topology framework. Taking the main deck beam, secondary deck beam, side bulkhead, top bulkhead and auxiliary components as examples, the gradient descent path for each step is planned using the potential energy field, with a step size of 5 mm / step and the path smoothness error controlled within ≤0.1 mm. The residuals of each interface are calculated and the compensation amount is decomposed. The compensation amount of the side bulkhead and deck interface is in the range of 0.2-0.5 mm, forming the initial assembly strategy.

[0047] The total adjustment cost and closure accuracy are used as evaluation indicators. The total adjustment cost includes time and energy consumption factors. The initial sequence has a total adjustment cost of 9.2 hours and an interface matching degree of 96.8%. Based on the evaluation results, the assembly sequence and corresponding path and compensation amount are updated iteratively. The compensation amount allocation logic is optimized in each iteration.

[0048] The convergence conditions were set as follows: the total adjustment cost fluctuation ≤3% for three consecutive iterations and the interface matching degree ≥98.5%. After 15 iterations, the convergence requirements were met, and the optimal assembly sequence was output. The adjusted sequence was "deck main beam, side bulkhead, deck secondary beam, top bulkhead and auxiliary components". The total time for the corresponding pose adjustment path was 7.3h, the average interface compensation was 0.32mm, and the closing interface matching degree reached 99.1%.

[0049] The above optimal assembly scheme is converted into motion instructions and process parameters that can drive the KUKA KR C4 robotic arm. The motion instructions cover rotation angle and movement trajectory, and the process parameters include welding compensation area and cutting amount, forming 128 standardized operation instructions, which constitute the optimal assembly strategy instruction set and can be directly used for on-site assembly operations.

[0050] S3. Convert the optimal assembly strategy instruction set into control instructions for the actuator, drive the actuator to dynamically adjust the ship's structural attitude, and simultaneously collect the actual measurement data after dynamic adjustment and feed it back to the data fusion processing in step S1 to update the dynamic accuracy field.

[0051] like Figure 2 As shown, according to another embodiment of the present invention, a precision control system for a modular ship structure is provided, the system comprising: The precision field construction module 1 is used to collect multi-source data of the ship structure in real time during the pre-assembly stage, and to perform spatiotemporal alignment and fusion processing on the multi-source data to generate a dynamic precision field that characterizes the deviation between the actual pose and the expected pose of the ship structure. Assembly strategy optimization module 2 is used to optimize the assembly sequence, posture adjustment path and interface compensation of the ship structure based on the dynamic precision field with the goal of achieving the best matching degree of the joint interface. It generates the optimal assembly strategy instruction set to guide the assembly operation of the ship structure. Assembly execution module 3 is used to convert the optimal assembly strategy instruction set into control instructions for the actuator, drive the actuator to dynamically adjust the ship structure attitude, and collect the actual measurement data after dynamic adjustment and feed it back to the data fusion processing process in the accuracy field construction module 1 to realize the update of the dynamic accuracy field.

[0052] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores static and dynamic information data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the above method embodiments.

[0053] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0054] In addition, the present invention 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 in the above method embodiments.

[0055] In addition, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0056] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for precision control of modular ship structures, characterized in that, The method includes: S1. Real-time acquisition of multi-source data of the ship structure during the pre-assembly stage, and spatiotemporal alignment and fusion processing of the multi-source data to generate a dynamic accuracy field to characterize the deviation between the actual pose and the desired pose of the ship structure. S2. With the goal of achieving the optimal matching degree of the main section closure interface, the assembly sequence, posture adjustment path and interface compensation amount of the ship structure are collaboratively optimized based on the dynamic precision field to generate the optimal assembly strategy instruction set to guide the assembly operation of the ship structure. S3. Convert the optimal assembly strategy instruction set into control instructions for the actuator, drive the actuator to dynamically adjust the ship's structural attitude, and simultaneously collect the actual measurement data after dynamic adjustment and feed it back to the data fusion processing in step S1 to update the dynamic accuracy field.

2. The method for precision control of modular ship structures according to claim 1, characterized in that, The real-time acquisition of multi-source data of the ship structure during the pre-assembly stage, followed by spatiotemporal alignment and fusion processing of the multi-source data to generate a dynamic accuracy field characterizing the deviation between the actual and desired pose of the ship structure, includes: S11. Integrate multi-source data of the ship structure during the pre-assembly stage collected by sensing devices, and perform timestamp synchronization and unified coordinate system processing on the multi-source fused data to generate an original pose observation sequence with consistent spatiotemporal reference. S12. The original pose observation sequence is associated and matched with the preset desired pose to obtain the position deviation features of each feature point. The position deviation features of each feature point are fused to generate the fused pose estimate of the key feature points on the surface of the ship structure. S13. Based on the fusion pose estimation of key feature points on the ship structure surface, the pose deviation of the entire ship structure is interpolated to generate a dynamic accuracy field containing spatial distribution deviation estimation.

3. The method for precision control of modular ship structures according to claim 2, characterized in that, The step of associating and matching the original pose observation sequence with the preset desired pose to obtain the position deviation features of each feature point, and then fusing the position deviation features of each feature point to generate a fused pose estimate of the key feature points on the ship structure surface includes: S121. Extract the observation coordinates of each observation feature point from the original pose observation sequence, and preliminarily correlate them with the corresponding theoretical feature point coordinates in the preset desired pose. S122. Based on the initial correlation features between feature points, the optimal spatial transformation parameters are obtained by registering the observed feature point set and the theoretical feature point set using the iterative nearest point algorithm. S123. Calculate the pose deviation of each observed feature point according to the optimal spatial transformation parameters, assign weights to each pose deviation based on preset rules and perform weighted fusion to obtain the fused pose estimate of the key feature points on the ship structure surface.

4. The method for precision control of modular ship structures according to claim 3, characterized in that, The process of registering the observed feature point set with the theoretical feature point set using the iterative nearest point algorithm to obtain the optimal spatial transformation parameters includes: S1221. Search for the nearest neighbor of each observed feature point in the theoretical feature point set to establish an initial corresponding point pair, and filter the initial corresponding point pairs in combination with the observation uncertainty to form a candidate corresponding point pair set. S1222. Based on the reciprocal of the observation uncertainty of each point pair in the candidate corresponding point pair set as the weight, construct and solve the weighted least squares problem, and calculate the optimal spatial transformation parameters that minimize the sum of squared weighted distances between all corresponding point pairs. S1223. Update the pose of the observed feature point set using the optimal spatial transformation parameters, and calculate the average matching error between the updated observed feature point set and the theoretical feature point set. S1224. If the average matching error is less than the preset threshold or the maximum number of iterations is met, the final optimal spatial transformation parameters are output. Otherwise, the pose of the updated observation feature point set is used as input and the process returns to step S1221 for the next iteration calculation until the average matching error is less than the preset threshold or the maximum number of iterations is reached.

5. The method for precision control of modular ship structures according to claim 4, characterized in that, The optimal assembly strategy instruction set, aimed at achieving the best matching degree of the main assembly interface, is generated by collaboratively optimizing the assembly sequence, pose adjustment path, and interface compensation amount of the ship structure based on a dynamic precision field. This optimization guides the ship structure assembly operation. S21. With the goal of achieving the optimal matching degree of the overall segment closure interface, a multi-objective optimization model is constructed based on the dynamic precision field. S22. The multi-objective optimization model is solved by combining coupled topological sorting and potential field-guided collaborative optimization strategy, and the optimal assembly scheme including the assembly sequence of ship structure, pose adjustment path and interface compensation amount is output. S23. Transform the optimal assembly scheme into action instructions and process parameters that can drive the actuator, forming an optimal assembly strategy instruction set to guide the assembly operation of the ship structure.

6. The method for precision control of modular ship structures according to claim 5, characterized in that, The combined optimization strategy of coupled topological sorting and potential field guidance is used to solve the multi-objective optimization model, and the output includes the optimal assembly scheme containing the assembly sequence of the ship structure, the pose adjustment path, and the interface compensation amount. S221. Based on the geometric and technological constraints of the ship structure, construct the assembly constraint topology with the dependency relationship of assembly sequence, and simultaneously generate the pose energy field to guide pose adjustment. S222. Generate candidate assembly sequences within the framework of assembly constraint topology. Utilize the pose energy field to plan the gradient descent path and interface compensation amount for pose adjustment at each assembly step in the candidate assembly sequence, thus forming the current assembly strategy. S223. Evaluate the quality of the current assembly strategy using the total adjustment cost and closing accuracy as indicators, and iteratively update the assembly sequence and its corresponding pose adjustment optimization path and interface compensation amount based on the evaluation results. S224. When the iteration meets the convergence condition, output the assembly sequence, pose adjustment path and interface compensation amount that make the overall segment closing interface matching degree optimal, as the optimal assembly scheme obtained by solving.

7. The method for precision control of modular ship structures according to claim 6, characterized in that, The gradient descent path and interface compensation amount for planning pose adjustment for each assembly step in the candidate assembly sequence using the pose energy field include: For each assembly step in the candidate assembly sequence, the current pose of the ship structure is set as the starting point of the path. Guided by the gradient descent direction of the pose energy field, several gradient descent paths for pose adjustment to reach the target assembly area are planned, and the pose at the end of the path is recorded. Calculate the residual between the end-path pose and the preset closing pose, decompose the residual into the part eliminated by the actuator and the part that needs to be compensated for by materials, and output the interface compensation amount required for the current assembly step.

8. A precision control system for a modular ship structure, used to implement the precision control method for a modular ship structure as described in any one of claims 1-7, characterized in that, The system includes: The precision field construction module is used to collect multi-source data of the ship structure in real time during the pre-assembly stage, and to perform spatiotemporal alignment and fusion processing on the multi-source data to generate a dynamic precision field that characterizes the deviation between the actual pose and the expected pose of the ship structure. The assembly strategy optimization module is used to optimize the assembly sequence, posture adjustment path and interface compensation of the ship structure based on the dynamic precision field with the goal of achieving the best matching degree of the joint interface. It generates the optimal assembly strategy instruction set to guide the assembly operation of the ship structure. The assembly execution module is used to convert the optimal assembly strategy instruction set into control instructions for the actuator, drive the actuator to dynamically adjust the ship's structural attitude, and collect the actual measurement data after dynamic adjustment and feed it back to the data fusion processing process in the accuracy field construction module to update the dynamic accuracy field.

9. 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 7.

10. 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 7.

Images