Submarine pipeline laying intelligent path planning system based on deep learning

Abstract

The invention relates to a submarine pipeline laying intelligent path planning system based on deep learning. The system comprises a sea area topographic feature extraction subsystem used for obtaining topographic data of a target seabed area and generating a sea area topographic feature map of the target seabed area based on the topographic data; the global laying path planning subsystem is used for generating an initial pipeline network layout scheme between a preset pipeline laying starting point and a preset pipeline laying ending point by adopting a graph search algorithm according to the sea area topographic feature graph; and the local laying path planning subsystem is used for calculating a laying complexity parameter of a connection path in the initial pipeline network layout scheme based on a deep learning model, and performing iterative adjustment on node coordinate information and path topology information by adopting a multi-objective optimization algorithm to generate a submarine pipeline laying intelligent path. By adopting the system, full-process automation, multi-target optimal solution and dynamic adaptation of submarine pipeline laying path planning can be realized.

Description

Technical Field

[0001] The present invention belongs to the field of information technology, and in particular relates to an intelligent path planning system for submarine pipeline laying based on deep learning. Background Art

[0002] In the field of submarine pipeline laying projects, pipeline laying path planning is a key link to ensure the smooth implementation of the project and improve the project efficiency and safety. However, when laying pipelines in complex submarine environments, there are often many technical challenges.

[0003] With the continuous development of marine resources, the demand for laying submarine pipelines is growing. However, the submarine environment is extremely complex, with complex and diverse topography, including various slope changes and obstacles. Accurately obtaining and analyzing submarine terrain information has become a major problem in path planning. In the past, it was difficult to obtain characteristic maps based on terrain data that accurately reflected the actual situation of the submarine terrain, and the ability to identify submarine obstacles was limited, which seriously affected the accuracy and reliability of path planning.

[0004] In terms of path planning algorithms, traditional methods also have limitations. Previous path planning algorithms have difficulty making full use of terrain feature information when dealing with complex seabed terrain. The initial pipeline network layout solutions generated are often not reasonable enough and cannot effectively avoid complex terrain areas, increasing construction difficulty and cost. In addition, traditional technologies lack effective algorithms and models, and have low data analysis capabilities, making it difficult to generate pipeline laying paths that meet actual engineering needs. Summary of the invention

[0005] Based on this, it is necessary to provide a deep learning-based intelligent path planning system for submarine pipeline laying that can realize automated, multi-objective, and dynamically adaptive path planning for submarine pipeline laying in response to the above-mentioned technical problems.

[0006] The present application provides a deep learning-based intelligent path planning system for submarine pipeline laying, including: A sea area terrain feature extraction subsystem is used to obtain terrain data of the target seabed area and generate a sea area terrain feature map of the target seabed area based on the terrain data. The terrain data includes slope characteristics and obstacle distribution characteristics. The global laying path planning subsystem is used to generate an initial pipeline network layout plan between the preset pipeline laying starting point and end point according to the sea area terrain feature map using a graph search algorithm. The initial pipeline network layout plan includes node coordinate information of the connection nodes and path topology information of the connection path; The local laying path planning subsystem is used to calculate the laying complexity parameters of the connection paths in the initial pipeline network layout plan based on the deep learning model, and use a multi-objective optimization algorithm to iteratively adjust the node coordinate information and path topology information to generate an intelligent path for submarine pipeline laying. The optimization objectives of the multi-objective optimization algorithm include the laying complexity parameters.

[0007] The above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying obtains terrain data including slope characteristics and obstacle distribution characteristics through the sea area terrain feature extraction subsystem, so that the system can comprehensively and meticulously grasp the terrain conditions of the target seabed area, accurately present the complexity of the terrain and the distribution of obstacles, and provide accurate and critical basic information for subsequent path planning; Through the global laying path planning subsystem, the graph search algorithm is used to plan the initial pipeline network layout plan, which can make the submarine pipeline laying path planning process fully analyze the submarine terrain characteristics, effectively avoid complex terrain areas and obstacle areas, and improve the rationality and feasibility of the initial plan; The local laying path planning subsystem uses a deep learning model to calculate the laying complexity parameters, which can integrate various factors that affect the difficulty of submarine pipeline laying and improve the practicality of path planning. It also uses a multi-objective optimization algorithm to iteratively adjust the node coordinate information and path topology information, which can take into account multiple preset goals at the same time, thereby generating an intelligent path that better meets actual engineering needs, thereby improving the overall quality and comprehensive benefits of submarine pipeline laying path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0009] Figure 1 A schematic diagram of the structure of an intelligent path planning system for laying submarine pipelines based on deep learning provided in one embodiment of the present application; Figure 2 A schematic diagram of the structure of a local paving path planning subsystem provided in one embodiment of the present application; Figure 3 A schematic diagram of the structure of another local paving path planning subsystem provided by an embodiment of the present application; Figure 4 A schematic diagram of the structure of a sea area terrain feature extraction subsystem provided in one embodiment of the present application; Figure 5A schematic diagram of the structure of another sea area terrain feature extraction subsystem provided in one embodiment of the present application; Figure 6 A schematic diagram of the structure of another deep learning-based intelligent path planning system for submarine pipeline laying provided as an embodiment of the present application. DETAILED DESCRIPTION

[0010] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0011] In an exemplary embodiment, Figure 1 As shown, a deep learning-based intelligent path planning system 100 for laying submarine pipelines is provided, comprising: The sea area terrain feature extraction subsystem 101 can be used to obtain terrain data of a target seabed area, and can generate a sea area terrain feature map of the target seabed area based on the terrain data. The terrain data may include slope features and obstacle distribution features.

[0012] Optionally, the sea area terrain feature extraction subsystem 101 can obtain acoustic detection terrain data of the target seabed area based on acoustic detection equipment, can obtain remote sensing detection terrain data of the target seabed area based on remote sensing detection equipment, and can perform data matching and data fusion on the acoustic detection terrain data and remote sensing detection terrain data after preprocessing to generate terrain data of the target seabed area. Among them, the acoustic detection equipment can include but is not limited to single-beam echo sounder, multi-beam echo sounder, coherent multi-beam system and side-scan sonar equipment; remote sensing detection equipment can include but is not limited to ship remote sensing equipment, aviation remote sensing equipment and aerospace remote sensing equipment; preprocessing can include but is not limited to radiation correction, geometric correction, filtering denoising, signal enhancement, image enhancement and format conversion.

[0013] Optionally, the sea area terrain feature extraction subsystem 101 can generate a three-dimensional terrain model of the target seabed area based on the terrain data of the target seabed area including slope characteristics and obstacle distribution characteristics, combined with point cloud processing technology, and construct a sea area terrain feature map based on the three-dimensional terrain model of the target seabed area.

[0014] Furthermore, the sea area terrain feature extraction subsystem 101 can reconstruct the surface of the point cloud data by local weighted least squares fitting based on the moving least squares method. The fitting function of the point cloud data in the local neighborhood can be expressed as: ; In the formula, is the fitting function, is the target point, is the order of the fitting function, is the coefficient to be determined, The sea area terrain feature extraction subsystem 101 can solve the unknown coefficients by minimizing the local weighted error function, and the minimization of the local weighted error function can be expressed as: ; In the formula, To minimize the local weighted error function, Target point The number of points in the neighborhood of is the weight function, Target point In the neighborhood of Points, is the first The height value of a point.

[0015] Optionally, the sea area terrain feature extraction subsystem 101 can project the slope features and obstacle distribution features into a two-dimensional terrain model of the target seabed area based on the terrain data of the target seabed area including the slope features and obstacle distribution features, and can construct a sea area terrain feature map based on the two-dimensional terrain model of the target seabed area.

[0016] For example, the topographic data of the seabed area may include elevation data. The sea area topographic feature extraction subsystem 101 may calculate the slope based on the elevation data, and the slope calculation formula may be: ; In the formula, is the slope, and Elevation exist Direction and The sea area terrain feature extraction subsystem 101 can convert terrain data from a geographic coordinate system into a plane coordinate system, and integrate slope data and obstacle distribution characteristics into the plane coordinate system to generate a two-dimensional terrain model of the target seabed area, and can construct a sea area terrain feature map based on the two-dimensional terrain model.

[0017] Schematically, the format of the sea area topographic feature map matches the input format of the graph search algorithm.

[0018] The global laying path planning subsystem 102 may be used to generate an initial pipeline network layout plan between a preset pipeline laying starting point and an end point according to a sea area terrain feature map by using a graph search algorithm. The initial pipeline network layout plan may include node coordinate information of connection nodes and path topology information of connection paths; Schematically, the global paving path planning subsystem 102 can be connected to the sea area terrain feature extraction subsystem 101 to obtain the sea area terrain feature map of the target seabed area generated by the sea area terrain feature extraction subsystem 101 and terrain data including slope characteristics and obstacle distribution characteristics.

[0019] Optionally, the global laying path planning subsystem 102 can convert the sea area terrain feature map into the input format of the graph search algorithm; use the graph search algorithm to set the starting point and the end point, initialize the graph model; use the graph search algorithm to calculate the shortest path; save the node coordinate information and path topology information of the path, and generate the initial pipeline network layout. Among them, the graph search algorithm can include but is not limited to the Dijkstra algorithm, the A* algorithm, the breadth-first search algorithm and the depth-first search algorithm. The input format of the graph search algorithm can be an adjacency list, and the path topology information includes the connection relationship between nodes.

[0020] The local laying path planning subsystem 103 can be used to calculate the laying complexity parameters of the connection paths in the initial pipeline network layout plan based on the deep learning model, and use a multi-objective optimization algorithm to iteratively adjust the node coordinate information and path topology information to generate an intelligent path for submarine pipeline laying. The optimization objectives of the multi-objective optimization algorithm include the laying complexity parameters.

[0021] Schematically, the local paving path planning subsystem 103 can be connected to the sea area terrain feature extraction subsystem 101 and the global paving path planning subsystem 102, obtain the sea area terrain feature map of the target seabed area generated by the sea area terrain feature extraction subsystem 101 and the terrain data including slope characteristics and obstacle distribution characteristics, and obtain the initial pipeline network layout plan including node coordinate information of the connection nodes and path topology information of the connection path generated by the global paving path planning subsystem 102.

[0022] Furthermore, the local laying path planning subsystem 103 can use the laying complexity parameter as the input of the multi-objective optimization algorithm, iteratively adjust the node coordinate information and the path topology information, generate the multi-objective optimization iterative output path, and perform secondary optimization on the multi-objective optimization iterative output path to reduce the number of turns and the turning range, and generate the intelligent path for laying the submarine pipeline. Among them, the input of the multi-objective optimization algorithm can be one or more of the laying complexity parameter, the pipeline length parameter, the obstacle distance parameter, the slope change parameter, the cost target parameter, the path smoothing parameter and the risk assessment parameter. The multi-objective optimization algorithm can include but is not limited to a multi-objective particle swarm optimization, a multi-objective genetic algorithm, a multi-objective ant colony optimization algorithm, a multi-objective non-dominated sorting genetic algorithm and a hybrid multi-objective optimization algorithm.

[0023] Optionally, the local laying path planning subsystem 103 can input the sea area terrain feature map, terrain data and the initial pipeline network layout plan of the target seabed area into the deep learning model to calculate the laying complexity parameters of the connection path in the initial pipeline network layout plan. The deep learning model can be constructed based on one or more models of a convolutional neural network model, a converter model and a long short-term memory network model.

[0024] Exemplarily, the input of the deep learning model can be a sea area terrain feature map and an initial pipeline network layout plan. The deep learning model can include a convolutional neural network module. The fully connected layer of the convolutional neural network module can be built based on the BP neural network. The output of the fully connected layer of the convolutional neural network module built based on the BP neural network can be used as the output of the deep learning model.

[0025] In the above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying, the accuracy and feasibility of path planning can be improved by acquiring terrain data including slope characteristics and obstacle distribution characteristics and generating corresponding sea area terrain feature maps; by adopting a graph search algorithm, possible connection paths can be quickly found under complex terrain conditions, thereby improving planning efficiency and reducing construction difficulty and cost; by using a deep learning model and adopting a multi-objective optimization algorithm, a variety of influencing factors can be comprehensively considered to realize the intelligence and automation of path planning, reduce manual intervention, improve work efficiency, and thus improve the quality of submarine pipeline laying path planning.

[0026] In an optional embodiment of the present application, please refer to Figure 2 The local paving path planning subsystem 103 may include a deep learning feature calculation module 201 , a paving complexity parameter calculation module 202 and a path iteration adjustment module 203 .

[0027] The deep learning feature calculation module 201 can be used to analyze the initial pipeline network layout plan based on the deep learning model according to the slope characteristics and obstacle distribution characteristics, and generate terrain distribution parameters and laying cost parameters. The terrain distribution parameters can be used to characterize the terrain undulation of the connection path, and the laying cost parameters can be used to reflect the laying cost of the connection path.

[0028] Schematically, the deep learning feature calculation module 201 can be connected to the sea terrain feature extraction subsystem and the global laying path planning subsystem to obtain the sea terrain feature map of the target seabed area generated by the sea terrain feature extraction subsystem and the terrain data including the slope characteristics and the obstacle distribution characteristics, and obtain the initial pipeline network layout plan including the node coordinate information of the connection nodes and the path topology information of the connection path generated by the global laying path planning subsystem. The deep learning feature calculation module 201 can also be connected to the path iteration adjustment module 203, and obtain the iterative path generated by the path iteration adjustment module 203 by iteratively adjusting the node coordinate information and the path topology information based on the multi-objective optimization algorithm.

[0029] Optionally, the deep learning model can be trained by a training data set constructed based on a pipeline network layout plan based on samples of historical submarine pipeline laying sea areas, terrain data including slope characteristics and obstacle distribution characteristics, terrain distribution parameters and laying cost parameters. The input of the deep learning model can be the initial pipeline network layout plan and terrain data including slope characteristics and obstacle distribution characteristics, and the output of the deep learning model can be the terrain distribution parameters and laying cost parameters of the initial pipeline network layout plan. Among them, the deep learning model can include a multi-layer perceptron module (MLP), and the output of the multi-layer perceptron module can be the output terrain distribution parameters and laying cost parameters.

[0030] Furthermore, the input of the deep learning model can be an iterative path generated by iteratively adjusting the node coordinate information and path topology information based on a multi-objective optimization algorithm by the path iteration adjustment module 203, as well as terrain data including slope characteristics and obstacle distribution characteristics. The output of the deep learning model can be the terrain distribution parameters and paving cost parameters of the iterative path generated by iteratively adjusting the node coordinate information and path topology information based on a multi-objective optimization algorithm by the path iteration adjustment module 203.

[0031] Schematically, the initial pipeline network layout plan and terrain data can achieve location matching.

[0032] The laying complexity parameter calculation module 202 can be used to combine the fuzzy comprehensive laying difficulty evaluation model, based on the terrain distribution characteristics and obstacle influence factors, to calculate the laying complexity parameters. The laying complexity parameters can be used to characterize the influence of the pipe section length and terrain complexity on the laying path.

[0033] Exemplarily, the laying complexity parameter calculation module 202 can be connected to the deep learning feature calculation module 201 to obtain the terrain distribution features generated by the deep learning feature calculation module 201; the laying complexity parameter calculation module 202 can be connected to the global laying path planning subsystem to obtain the initial pipeline network layout plan generated by the global laying path planning subsystem; the laying complexity parameter calculation module 202 can be connected to the path iteration adjustment module 203 to obtain the iterative path generated by the path iteration adjustment module 203.

[0034] Optionally, the paving complexity parameter calculation module 202 may input the preset obstacle impact factor and the terrain distribution feature generated by the deep learning feature calculation module 201 into the fuzzy comprehensive paving difficulty evaluation model to calculate the paving complexity parameter.

[0035] Optional, factor set of fuzzy comprehensive paving difficulty evaluation model It may include but is not limited to terrain distribution and obstacle density, and a set of comments for the fuzzy comprehensive paving difficulty evaluation model. It may include but is not limited to low, lower, medium, higher and high. The fuzzy relationship matrix of the fuzzy comprehensive paving difficulty evaluation model may include Factors For Level Comments The degree of membership The fuzzy comprehensive paving difficulty evaluation model can be based on the factor set Get the corresponding weight vector , and calculate the evaluation result vector based on the weight vector and fuzzy relationship matrix Among them, the calculation formula of the evaluation result vector can be: ; In the formula, The result vector is item, is the dimension of the factor set, is the weight vector The fuzzy comprehensive paving difficulty evaluation model can be based on the comment set Get the score vector corresponding to each level The calculation formula of the laying complexity parameter can be: ; In the formula, is the laying complexity parameter, is the number of levels of the comment set, is the score vector item.

[0036] The path iteration adjustment module 203 is used to iteratively adjust the node coordinate information and the path topology information based on the multi-objective optimization algorithm to generate an intelligent path for laying the submarine pipeline. The optimization objectives of the multi-objective optimization algorithm also include laying cost parameters.

[0037] Exemplarily, the path iteration adjustment module 203 may be connected to the laying complexity parameter calculation module 202 to obtain the laying complexity parameter generated by the laying complexity parameter calculation module 202; the path iteration adjustment module 203 may be connected to the deep learning feature calculation module 201 to obtain the laying cost parameter generated by the deep learning feature calculation module 201. The path iteration adjustment module 203 may use the laying cost parameter and the laying complexity parameter as optimization targets of the multi-objective optimization algorithm.

[0038] In the above-mentioned submarine pipeline laying intelligent path planning system based on deep learning, the slope characteristics and obstacle distribution characteristics can be deeply analyzed based on the deep learning model. Compared with traditional technical solutions, it can more accurately grasp the complex relationship between the initial pipeline network layout plan and the terrain and obstacles, thereby generating terrain distribution parameters and laying cost parameters that are more in line with the actual terrain conditions, providing a more reliable data basis for subsequent path planning; by adopting a multi-objective optimization algorithm, the node coordinate information and path topology information are iteratively adjusted, and a variety of relevant factors can be taken into account to generate a submarine pipeline laying intelligent path with higher comprehensive benefits, which can effectively reduce the total cost of pipeline laying, improve resource utilization efficiency, and ensure the quality and safety of pipeline laying; by adopting deep learning technology and fuzzy control technology, the formulation of pipeline laying plans can be made more intelligent and automated, reducing errors caused by manual intervention and subjective judgment, and improving the work efficiency and scientificity of decision-making in submarine pipeline laying path planning.

[0039] In an optional embodiment of the present application, please refer to Figure 3 The local laying path planning subsystem 103 also includes a submarine terrain commonality migration module 310, which includes a commonality feature extraction component 311, a feature label generation component 312, and a deep learning model adjustment component 313, wherein: The common feature extraction component 311 can be used to extract the common features of the sea area terrain feature map according to the slope features and obstacle distribution features. The common features are used to characterize the topographic commonalities between the target seabed area and the sea area where the historical submarine pipeline is laid.

[0040] Exemplarily, the common feature extraction component 311 may be connected to the sea area terrain feature extraction subsystem to obtain the terrain data of the target seabed area acquired by the sea area terrain feature extraction subsystem.

[0041] Optionally, the common feature extraction component 311 can convert the terrain data of the target seabed area and the historical seabed pipeline laying area into a unified data format, and perform data cleaning and normalization processing on the terrain data including the slope characteristics and the obstacle distribution characteristics. The common feature extraction component 311 can extract the common features of the slope characteristics and the obstacle distribution characteristics of the processed terrain data of the target seabed area and the historical seabed pipeline laying area to obtain common features. The common feature extraction can be one or more of correlation analysis, variance selection analysis, principal component analysis and Z value analysis.

[0042] The feature label generation component 312 can be used to input common features into the cluster analysis model, classify, and generate semantic feature labels. The semantic feature labels include continental shelf labels, continental slope labels, ocean basin labels, and trench labels.

[0043] Exemplarily, the feature label generation component 312 can be connected to the common feature extraction component 311 to obtain the common features extracted by the common feature extraction component 311. The clustering analysis model can be, but is not limited to, a K-Means clustering analysis model, a hierarchical clustering analysis model, a density-based noisy space clustering analysis model, and a Gaussian mixture probability-based clustering analysis model. The feature label generation component 312 can input the standardized common features into the selected clustering algorithm for classification, analyze the generated clustering results, and determine the semantic feature labels corresponding to each cluster.

[0044] Optionally, taking the K-Means clustering analysis model as an example for illustration, the feature label generation component 312 can: an initialization step of randomly initializing K cluster centers; an assignment step of assigning each sample to the cluster to which the nearest cluster center belongs; an update step of updating the cluster center to the mean of all samples in each cluster; and repeating the assignment step and the update step until the cluster center no longer changes or the maximum number of iterations is reached.

[0045] The deep learning model adjustment component 313 can be used to identify the target approximate historical submarine pipeline laying sea area in the historical submarine pipeline laying sea area based on semantic feature labels, and update the parameters of the deep learning model by improving the feature alignment loss function combined with the terrain data of the target seabed area and the terrain data of the target approximate historical submarine pipeline laying sea area.

[0046] Exemplarily, the deep learning model adjustment component 313 can be connected to the feature label generation component 312 and the deep learning feature calculation module 201 to adjust the deep learning model in the deep learning feature calculation module 201 based on the semantic feature labels generated by the feature label generation component 312.

[0047] Optionally, the expression of the improved feature alignment loss function can be: ; In the formula, To improve the feature alignment loss function, The number of samples of the sea area where the target approximates the historical submarine pipeline laying, is the number of samples in the target seafloor area, is the feature mapping function, For the The target is similar to the sample of the historical submarine pipeline laying area. For the samples from the target seafloor area, is the square of the norm in the reproducing kernel Hilbert space (RKHS).

[0048] The above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying can extract the common characteristics of terrain data according to the slope characteristics and obstacle distribution characteristics, effectively identify the similarities between the current sea area terrain and the historical submarine pipeline laying sea area, and provide a more accurate terrain information basis for subsequent path planning. By adopting corresponding laying strategies according to the characteristics of different terrain types in path planning, the pertinence and reliability of laying path planning can be improved.

[0049] Furthermore, the above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying updates the parameters of the deep learning model by improving the feature alignment loss function and combining the terrain data of the target submarine area and the terrain data of the target approximate historical submarine pipeline laying sea area, so that the deep learning model can better adapt to the pipeline laying path planning tasks under different submarine terrain conditions, which can improve the generalization ability and accuracy of the deep learning model, thereby further improving the performance and quality of path planning.

[0050] In an optional embodiment of the present application, the input of the improved feature alignment loss function is a feature mapping function of the target approximate historical submarine pipeline laying sea area and a feature mapping function of the target seabed area, and the feature mapping function is a function that maps the sea area terrain feature map to terrain distribution features and laying cost features.

[0051] Among them, the expression of the improved feature alignment loss function can be: ; In the formula, To improve the feature alignment loss function, is the coefficient of the maximum mean difference term of terrain distribution, The number of samples of the sea area where the target approximates the historical submarine pipeline laying, is the number of samples in the target seafloor area, is the terrain distribution subfunction of the feature mapping function, For the The target is similar to the sample of the historical submarine pipeline laying area. For the samples from the target seafloor area, is the square of the norm in the reproducing kernel Hilbert space (RKHS), is the cross-constraint coefficient of the laying cost, For the The cost characteristics of the samples in the target seafloor area are compared with the gradient of the terrain characteristics, is the target approximation history cost terrain gradient hyperparameter, The goal is to approximate the expected gradient of the cost characteristics of a sample of historical submarine pipeline laying areas with respect to the terrain characteristics.

[0052] In the above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying, by analyzing the target approximate historical submarine pipeline laying sea areas and target submarine area samples, it can effectively capture the differences and commonalities in terrain characteristics and cost characteristics of different areas, and help apply historical experience to path planning in the current target area. It can maximize the economy and safety of submarine pipeline laying while ensuring the quality of the project.

[0053] In an optional embodiment of the present application, please refer to Figure 3 The path iteration adjustment module 203 includes an improved sparrow search algorithm component 341 .

[0054] The sparrow search algorithm component 341 can be used to adjust the node coordinate information and the path topology information by using the improved sparrow search algorithm based on the laying complexity parameter and the laying cost parameter. The sparrow individuals of the improved sparrow search algorithm include the node coordinate information and the path topology information.

[0055] Among them, the finder position update formula of the improved sparrow search algorithm can be: ; In the formula, For the The iteration Only the finder's order from the starting point to the end point The node coordinates of the nodes, For the front The best one in the iteration is the first one in the order from the start to the end. The node coordinates of the nodes, For the The iteration Only the finder's order from the starting point to the end point The node coordinates, is the random weight coefficient, for The random number inside for The random number inside is the maximum number of iterations, is a random number that follows a normal distribution, is a matrix of the same type as the node coordinate matrix whose elements are all 1.

[0056] In the above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying, the node coordinates and path topology are adjusted by adopting an improved sparrow search algorithm, which can improve the global search capability and convergence speed compared with the traditional search algorithm.

[0057] In an optional embodiment of the present application, please refer to Figure 3 The local laying path planning subsystem 103 may include a submarine terrain commonality migration module 310, a deep learning feature calculation module 201, a laying complexity parameter calculation module 202, a path iteration adjustment module 203, and a submarine pipeline digital twin display module 350. The local laying path planning subsystem 103 can significantly improve the intelligent level of submarine pipeline laying, optimize the laying path, reduce construction risks and costs, and improve the overall quality of the project through the above modules.

[0058] In an optional embodiment of the present application, please refer to Figure 4 The sea area terrain feature extraction subsystem 101 includes a three-dimensional terrain model generation module 410 and a sea area terrain feature map generation module 420 .

[0059] The 3D terrain model generation module 410 can be used to construct a differentiated obstacle-annotated 3D terrain model based on terrain data. The differentiated obstacle-annotated 3D terrain model is used to annotate obstacle blocks in differentiated blocks whose sizes are adapted to the complexity of the terrain.

[0060] Optionally, the 3D terrain model generation module 410 can input terrain data into the 3D modeling model to construct an initial 3D terrain model, and perform block differentiation simplification on the generated initial 3D model based on the slope characteristics and obstacle distribution characteristics of the terrain data to generate a differentiated 3D terrain model. The 3D terrain model generation module 410 can perform obstacle annotation on the differentiated 3D terrain model to obtain a differentiated obstacle annotation 3D terrain model. The 3D modeling model can be, but is not limited to, a regular grid 3D modeling model, an irregular triangulated network 3D modeling model, and a point cloud 3D modeling model.

[0061] The sea area terrain feature map generation module 420 can be used to map the differentiated obstacle annotated three-dimensional terrain model to a two-dimensional plane, generate an obstacle annotated two-dimensional grid map, and construct a sea area terrain feature map based on the obstacle annotated two-dimensional grid map.

[0062] Exemplarily, the sea area terrain feature map generation module 420 can be connected to the three-dimensional terrain model generation module 410, obtain the differentiated obstacle annotated three-dimensional terrain model constructed by the three-dimensional terrain model generation module 410, and map the differentiated obstacle annotated three-dimensional terrain model to a two-dimensional plane to generate an obstacle annotated two-dimensional grid map, and then construct a sea area terrain feature map based on the obstacle annotated two-dimensional grid map.

[0063] In the above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying, by constructing a differentiated obstacle-annotated three-dimensional terrain model, it is possible to clearly mark the obstacle blocks in the differentiated blocks that adapt to the complexity of the terrain, and can intuitively reflect the three-dimensional structural characteristics of the submarine terrain and the distribution characteristics of the obstacles; by simplifying the three-dimensional spatial information into two-dimensional plane data, it is easy to simplify the complexity of data processing and analysis, and improve the system's data storage, retrieval and computing performance; the sea area terrain feature map constructed based on the obstacle-annotated two-dimensional grid map can integrate terrain and obstacle information, and improve the performance of subsequent graph search algorithms.

[0064] In an optional embodiment of the present application, please refer to Figure 4 , the three-dimensional terrain model generation module 410 includes a point cloud processing component 411, a block division component 412 and an obstacle annotation component 413, wherein: The point cloud processing component 411 can be used to construct a three-dimensional terrain model of the target seabed area based on terrain data in combination with point cloud processing technology.

[0065] Optionally, the point cloud processing component 411 can use professional ocean detection equipment such as multi-beam bathymetry systems and side-scan sonars to collect terrain data of the target seabed area and construct an original point cloud data set. The point cloud processing component 411 can filter and reduce noise and perform data registration on the original point cloud data set to obtain optimized point cloud data. The point cloud processing component 411 can use a surface reconstruction algorithm based on the optimized point cloud data to construct a three-dimensional terrain model of the target seabed area.

[0066] The block division component 412 can be used to calculate the terrain complexity parameter matrix based on the three-dimensional terrain model in combination with the slope characteristics and the obstacle distribution characteristics, and use an adaptive grid division algorithm to divide the three-dimensional terrain model into grid blocks whose sizes are proportional to the terrain complexity parameters based on the terrain complexity parameter matrix to construct a differentiated three-dimensional terrain model.

[0067] Optionally, the block division component 412 can be connected to the point cloud processing component 411, obtain the three-dimensional terrain model generated by the point cloud processing component 411, combine the slope characteristics and obstacle distribution characteristics of the terrain data of the target seabed area, calculate the complexity parameters of the terrain data of the target seabed area, and form a terrain complexity parameter matrix. The block division component 412 can use an adaptive grid division algorithm based on the terrain complexity parameter matrix to divide the three-dimensional terrain model into grid blocks whose sizes are proportional to the terrain complexity parameters, and construct a differentiated three-dimensional terrain model.

[0068] The obstacle annotation component 413 can be used to input the slope characteristics and obstacle distribution characteristics of the grid blocks of the differentiated three-dimensional terrain model into the three-dimensional obstacle recognition deep learning model, identify the obstacle grid blocks in the differentiated three-dimensional terrain model, and construct a differentiated obstacle annotated three-dimensional terrain model.

[0069] Optionally, the three-dimensional obstacle recognition deep learning model can be a three-dimensional convolutional neural model. The three-dimensional obstacle recognition deep learning model can identify obstacle grid blocks in the differentiated three-dimensional terrain model. The obstacle annotation component 413 can perform obstacle annotation on the identified obstacle grid blocks to construct a differentiated obstacle annotation three-dimensional terrain model.

[0070] In the above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying, a large number of terrain data points can be accurately collected and processed through point cloud processing technology, so that the true situation of the submarine terrain can be highly restored; the three-dimensional terrain model is divided into grid blocks whose size is proportional to the terrain complexity parameter by adopting an adaptive grid division algorithm, which can reduce unnecessary calculations while ensuring the accuracy and reliability of the system and improve the processing efficiency of the system; through the powerful learning and recognition capabilities of the three-dimensional obstacle recognition deep learning model, the obstacle grid blocks in the differentiated three-dimensional terrain model can be automatically and accurately identified, thereby improving the efficiency and accuracy of obstacle recognition.

[0071] In an optional embodiment of the present application, please refer to Figure 5 ,The sea terrain feature map can be used to annotate the two-dimensional raster map for obstacles.

[0072] The sea area terrain feature extraction subsystem 101 may include a grid map initialization module 501, a risk assessment module 502, and a two-dimensional obstacle annotation module 503, wherein: The grid map initialization module 501 can be used to map terrain data to a two-dimensional plane to generate a two-dimensional initial grid map. The grids of the two-dimensional initial grid map include slope characteristic attribute information and obstacle distribution characteristic attribute information.

[0073] The risk assessment module 502 may be used to calculate a risk assessment value of a grid based on the slope characteristic attribute information and the obstacle distribution characteristic attribute information.

[0074] The two-dimensional obstacle annotation module 503 can be used to input the risk assessment value of the grid into the two-dimensional obstacle recognition deep learning model, identify the obstacle grid in the two-dimensional initial grid map, and construct an obstacle annotated two-dimensional grid map.

[0075] Optionally, the two-dimensional obstacle recognition deep learning model may be a convolutional neural network model.

[0076] In the above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying, by mapping terrain data to a two-dimensional plane to generate a two-dimensional initial grid map, the planar expression of complex three-dimensional terrain data can be achieved, making the data easier to process and analyze; by converting terrain data into a two-dimensional grid map with specific attribute information, a unified data format can be provided for subsequent processing, which is convenient for subsequent modules to perform standardized processing and improve the compatibility and versatility of the system; by utilizing the powerful pattern recognition ability of the deep learning model to automatically and accurately identify obstacle grids in the two-dimensional initial grid map, compared with traditional manual recognition or simple rule judgment methods, it can handle complex terrain features and risk relationships, greatly improving the accuracy and efficiency of obstacle recognition.

[0077] In an optional embodiment of the present application, the graph search algorithm is an ant colony algorithm.

[0078] Optionally, the sea area terrain feature map is a grid map, and the global laying path planning subsystem can set the pheromone importance factor, the heuristic information importance factor, the pheromone volatilization rate and the pheromone update constant, and initialize the pheromone matrix of the sea area terrain feature map; the global laying path planning subsystem can use the preset pipeline laying starting point and end point as the starting node and target node of the sea area terrain feature map; the global laying path planning subsystem can control each ant to select the next node according to the calculated probability until it reaches the end node; the global laying path planning subsystem can update the pheromone concentration according to the volatilization rate and the path length of the ant, repeat the path selection and pheromone update until the termination condition is met, and generate an initial pipeline network layout plan.

[0079] In an optional embodiment of the present application, please refer to Figure 3 The local laying path planning subsystem may also include a submarine pipeline digital twin display module 350 .

[0080] The submarine pipeline digital twin display module 350 can be connected to the path iteration adjustment module 203 and the sea area terrain feature extraction subsystem, and generate and display the submarine pipeline digital twin model based on the submarine pipeline laying intelligent path generated by the path iteration adjustment module 203 and the sea area terrain feature map generated by the sea area terrain feature extraction subsystem.

[0081] In an exemplary embodiment, Figure 6 As shown, a deep learning-based intelligent path planning system 100 for laying submarine pipelines is provided, comprising: a sea area terrain feature extraction subsystem 101, a global laying path planning subsystem 102 and a local laying path planning subsystem 103, wherein: The sea area terrain feature extraction subsystem 101 may include a three-dimensional terrain model generation module 410 and a sea area terrain feature map generation module 420. The sea area terrain feature map generation module 420 may be connected to the three-dimensional terrain model generation module 410 and the global paving path planning subsystem 102.

[0082] The local laying path planning subsystem 103 may include a submarine terrain commonality migration module 310, a deep learning feature calculation module 201, a laying complexity parameter calculation module 202, a path iteration adjustment module 203, and a submarine pipeline digital twin display module 350. The submarine terrain commonality migration module 310 may be connected to the sea area terrain feature extraction subsystem 101 and the deep learning feature calculation module 201; the deep learning feature calculation module 201 may be connected to the global laying path planning subsystem 102, the submarine terrain commonality migration module 310, and the laying complexity parameter calculation module 202; the submarine pipeline digital twin display module 350 may be connected to the path iteration adjustment module 203 and the sea area terrain feature extraction subsystem 101.

[0083] In the above-mentioned deep learning-based intelligent path planning system for submarine pipeline laying, through the collaborative work of various subsystems and modules, it is possible to accurately process terrain data and scientifically plan the submarine pipeline laying path, thereby improving the efficiency, safety and economy of the submarine pipeline laying project.

[0084] The above-mentioned embodiments only express several implementation methods of the embodiments of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the embodiments of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the embodiments of the present application, and these all belong to the protection scope of the embodiments of the present application.

Claims

1. A deep learning-based intelligent path planning system for submarine pipeline laying, characterized in that: include: A sea area terrain feature extraction subsystem is used to obtain terrain data of a target seabed area and generate a sea area terrain feature map of the target seabed area based on the terrain data, wherein the terrain data includes slope characteristics and obstacle distribution characteristics; A global laying path planning subsystem, for generating an initial pipeline network layout plan between a preset pipeline laying starting point and an end point according to the sea area terrain feature map by using a graph search algorithm, wherein the initial pipeline network layout plan includes node coordinate information of connection nodes and path topology information of connection paths; The local laying path planning subsystem is used to calculate the laying complexity parameters of the connection paths in the initial pipeline network layout plan based on the deep learning model, and iteratively adjust the node coordinate information and the path topology information using a multi-objective optimization algorithm to generate an intelligent path for laying submarine pipelines, wherein the optimization objectives of the multi-objective optimization algorithm include the laying complexity parameters.

2. The system according to claim 1, characterized in that: The local paving path planning subsystem includes a deep learning feature calculation module, a paving complexity parameter calculation module and a path iteration adjustment module; The deep learning feature calculation module is used to analyze the initial pipeline network layout plan according to the slope characteristics and the obstacle distribution characteristics based on the deep learning model, and generate terrain distribution parameters and laying cost parameters, wherein the terrain distribution parameters are used to characterize the terrain undulation of the connection path, and the laying cost parameters are used to reflect the laying cost of the connection path; The laying complexity parameter calculation module is used to calculate the laying complexity parameter based on the terrain distribution characteristics and obstacle influence factors in combination with the fuzzy comprehensive laying difficulty evaluation model, and the laying complexity parameter is used to characterize the influence of the pipe section length and terrain complexity on the laying path; The path iteration adjustment module is used to iteratively adjust the node coordinate information and the path topology information based on the multi-objective optimization algorithm to generate the submarine pipeline laying intelligent path. The optimization target of the multi-objective optimization algorithm also includes the laying cost parameter.

3. The system according to claim 2, characterized in that: The local paving path planning subsystem also includes a seabed terrain commonality migration module, which includes a commonality feature extraction component, a feature label generation component, and a deep learning model adjustment component; The common feature extraction component is used to extract the common features of the terrain data according to the slope features and the obstacle distribution features, wherein the common features are used to characterize the commonalities of the terrain between the target seabed area and the sea area where the historical submarine pipeline was laid; The feature label generation component is used to input the common features into the cluster analysis model, classify them, and generate semantic feature labels, wherein the semantic feature labels include continental shelf labels, continental slope labels, ocean basin labels, and trench labels; The deep learning model adjustment component is used to identify the target approximate historical submarine pipeline laying sea area in the historical submarine pipeline laying sea area based on the semantic feature label, and update the parameters of the deep learning model by combining the terrain data of the target seabed area and the terrain data of the target approximate historical submarine pipeline laying sea area through an improved feature alignment loss function.

4. The system according to claim 3, characterized in that: The input of the improved feature alignment loss function is the feature mapping function of the target approximate historical submarine pipeline laying sea area and the feature mapping function of the target seabed area, and the feature mapping function is a function mapping the terrain data to the terrain distribution characteristics and the laying cost characteristics; Among them, the expression of the improved feature alignment loss function is: ； In the formula, To improve the feature alignment loss function, is the coefficient of the maximum mean difference term of terrain distribution, The number of samples of the sea area where the target approximates the historical submarine pipeline laying, is the number of samples in the target seafloor area, is the terrain distribution subfunction of the feature mapping function, For the The target is similar to the sample of the historical submarine pipeline laying area. For the samples from the target seafloor area, is the square of the norm in the reproducing kernel Hilbert space (RKHS), is the cross-constraint coefficient of the laying cost, For the The cost characteristics of the samples in the target seafloor area are compared with the gradient of the terrain characteristics, is the target approximation history cost terrain gradient hyperparameter, The goal is to approximate the expected gradient of the cost characteristics of a sample of historical submarine pipeline laying areas with respect to the terrain characteristics.

5. The system according to claim 2, characterized in that: The path iteration adjustment module includes an improved sparrow search algorithm component, which is used to: adjust the node coordinate information and the path topology information using an improved sparrow search algorithm based on the paving complexity parameter and the paving cost parameter, and the sparrow individual of the improved sparrow search algorithm includes the node coordinate information and the path topology information; Among them, the finder position update formula of the improved sparrow search algorithm is: ； In the formula, For the The iteration Only the finder's order from the starting point to the end point The node coordinates of the nodes, For the front The best one in the iteration is the first one in the order from the start to the end. The node coordinates of the nodes, is the first iteration The iteration Only the finder's order from the starting point to the end point The node coordinates, is the random weight coefficient, for The random number inside for The random number inside is the maximum number of iterations, is a random number that follows a normal distribution, The elements are all The node coordinate matrix of .

6. The system according to claim 1, characterized in that: The sea area terrain feature extraction subsystem includes a three-dimensional terrain model generation module and a sea area terrain feature map generation module: The three-dimensional terrain model generation module is used to construct a differentiated obstacle-annotated three-dimensional terrain model based on the terrain data, wherein the differentiated obstacle-annotated three-dimensional terrain model is used to annotate obstacle blocks in differentiated blocks whose sizes are adapted to the complexity of the terrain; The sea area terrain feature map generation module is used to map the differentiated obstacle-annotated three-dimensional terrain model to a two-dimensional plane, generate an obstacle-annotated two-dimensional grid map, and construct the sea area terrain feature map based on the obstacle-annotated two-dimensional grid map.

7. The system according to claim 6, characterized in that: The three-dimensional terrain model generation module includes a point cloud processing component, a block division component and an obstacle marking component: The point cloud processing component is used to construct a three-dimensional terrain model of the target seabed area based on the terrain data in combination with point cloud processing technology; The block division component is used to calculate a terrain complexity parameter matrix based on the three-dimensional terrain model in combination with the slope characteristics and the obstacle distribution characteristics, and to divide the three-dimensional terrain model into grid blocks whose sizes are proportional to the terrain complexity parameters based on the terrain complexity parameter matrix using an adaptive grid division algorithm to construct a differentiated three-dimensional terrain model; The obstacle annotation component is used to input the slope characteristics and the obstacle distribution characteristics of the grid blocks of the differentiated three-dimensional terrain model into the three-dimensional obstacle recognition deep learning model, identify the obstacle grid blocks in the differentiated three-dimensional terrain model, and construct a differentiated obstacle annotated three-dimensional terrain model.

8. The system according to claim 1, characterized in that: The sea area topographic feature map is a two-dimensional grid map with obstacle markings; The sea area terrain feature extraction subsystem includes a grid map initialization module, a risk assessment module and a two-dimensional obstacle marking module: A grid map initialization module, used for mapping the terrain data to a two-dimensional plane to generate a two-dimensional initial grid map, wherein the grids of the two-dimensional initial grid map include slope characteristic attribute information and obstacle distribution characteristic attribute information; The risk assessment module is used to calculate the risk assessment value of the grid based on the slope characteristic attribute information and the obstacle distribution characteristic attribute information; The two-dimensional obstacle annotation module is used to input the risk assessment value of the grid into the two-dimensional obstacle recognition deep learning model, identify the obstacle grid in the two-dimensional initial grid map, and construct the obstacle annotated two-dimensional grid map.

9. The system according to any one of claims 6 to 8, characterized in that: The graph search algorithm is an ant colony algorithm.

10. The system according to any one of claims 1 to 8, characterized in that: The local laying path planning subsystem also includes a submarine pipeline digital twin display module; The submarine pipeline digital twin display module is used to generate and display a submarine pipeline digital twin model based on the submarine pipeline laying intelligent path and the sea area terrain feature map.

Images