An underwater terrain survey method and system based on unmanned ship data collection

Abstract

The present invention discloses an underwater terrain survey method and system based on unmanned ship data collection, which relates to the technical field of underwater terrain survey. It includes generating a dynamic environment map based on the collected multi-modal environment data; based on the dynamic environment map, using the deep Q-learning algorithm to dynamically perceive the underwater environment and predict the probability of potential terrain changes; judging whether to update the dynamic environment map according to the predicted probability of potential terrain changes, and planning a navigation path based on the updated dynamic environment map using the RRT algorithm; collecting terrain data based on the navigation path and generating a three-dimensional terrain model, and using U-Net to detect the shadow areas in the three-dimensional terrain model; according to the detected shadow areas, generating supplementary collection points through geometric analysis, adjusting the navigation path according to the supplementary collection points, and performing supplementary collection on the shadow parts to obtain supplementary collection data; fusing the supplementary collection data with the three-dimensional terrain model to generate a terrain survey report. The present invention improves the efficiency and accuracy of underwater terrain survey through dynamic environment perception, path planning, shadow area detection and supplementary collection.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater terrain surveying, and particularly to an underwater terrain surveying method and system based on data collection by an unmanned vessel. Background Art

[0002] With the continuous growth of underwater resource development, environmental protection, and national defense needs, underwater terrain surveying technology has developed rapidly. Traditional underwater terrain surveys usually rely on manned vessels or sonar systems at fixed stations, but there are problems such as low efficiency, limited coverage, and high costs. As a new type of intelligent surveying platform, an unmanned surface vehicle (USV) has advantages such as high automation, flexible operation, and the ability to operate in dangerous or complex waters, and has gradually become an important tool for underwater terrain surveying. Combining the development of multi-modal sensors (such as multi-beam sonar, side-scan sonar, etc.) with artificial intelligence algorithms, unmanned vessels have significant potential in improving the efficiency and accuracy of underwater terrain surveying. However, there are still deficiencies in dynamic path planning and data collection integrity in the existing technology, and optimization is urgently needed.

[0003] The main problems of the existing technology are as follows: First, the underwater environment is complex and changeable, and existing path planning algorithms mostly rely on static environment data, making it difficult to respond to obstacles or water flow changes in a timely manner, resulting in reduced survey efficiency or even mission failure; second, in terrain modeling, due to terrain occlusion or sensor blind spots, there are often uncovered shadow areas, affecting the integrity of the model. Although some technologies attempt to solve this problem by improving sensor resolution or increasing the acquisition density, there is a lack of an effective method that combines real-time path optimization and a supplementary acquisition mechanism, making it difficult to balance efficiency and accuracy. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an underwater terrain surveying method based on data collection by an unmanned vessel to solve the problems of insufficient adaptability in dynamic path planning and missing data in terrain modeling.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides an underwater terrain survey method based on unmanned ship data collection, which includes generating a dynamic environment map based on the collected multi-modal environment data; based on the dynamic environment map, using the deep Q-learning algorithm to dynamically perceive the underwater environment and predict the probability of potential terrain changes; judging whether to update the dynamic environment map according to the predicted probability of potential terrain changes, and planning a navigation path based on the updated dynamic environment map using the RRT algorithm; collecting terrain data based on the navigation path and generating a three-dimensional terrain model, and using U-Net to detect the shadow areas in the three-dimensional terrain model; according to the detected shadow areas, generating supplementary collection points through geometric analysis, adjusting the navigation path according to the supplementary collection points, performing supplementary collection on the shadow parts to obtain supplementary collection data; fusing the supplementary collection data with the three-dimensional terrain model to generate a terrain survey report.

[0008] As a preferred embodiment of the underwater terrain survey method based on unmanned ship data collection according to the present invention, wherein: the steps of generating a dynamic environment map based on the collected multi-modal environment data are as follows:

[0009] Collect multi-modal environment data in real time through sensors; the multi-modal environment data includes underwater terrain, obstacle distribution, and water flow velocity;

[0010] Use Bayesian fusion to fuse the multi-modal environment data, and combine GPS to perform spatial positioning on the multi-modal environment data, and generate the navigation trajectory of the unmanned ship according to the spatial positioning;

[0011] Based on the navigation trajectory of the unmanned ship and the multi-modal environment data, generate a dynamic environment map through Delaunay.

[0012] As a preferred embodiment of the underwater terrain survey method based on unmanned ship data collection according to the present invention, wherein: the steps of dynamically perceiving the underwater environment and predicting the probability of potential terrain changes based on the dynamic environment map using the deep Q-learning algorithm are as follows:

[0013] Through feature extraction and discretization methods, convert the data features in the dynamic environment map into the state space of the deep Q-learning algorithm, and use the data features of the current local area of the underwater environment as the state in the state space;

[0014] Define the action space based on the operations of the unmanned ship in the current environment as actions;

[0015] Calculate the comprehensive reward value based on the distance between the unmanned ship and the obstacle, the coverage ratio of the target area, and the navigation cost ;

[0016] Based on the comprehensive reward value , predicting the potential terrain change probability by combining the divergence degree of the water flow velocity field, the terrain depth, and the obstacle density in the underwater environment .

[0017] As a preferred solution of the underwater terrain survey method based on unmanned ship data collection according to the present invention, wherein: judging whether to update the dynamic environment map according to the predicted potential terrain change probability, and planning the navigation path by using the updated dynamic environment map based on the RRT algorithm, the specific steps are as follows:

[0018] Defining a threshold terrain change risk threshold C1 based on the historical terrain change frequency and change amplitude;

[0019] When ≥ C1, it is considered that the potential change probability of the terrain in the current area is relatively high, and the dynamic environment map is updated;

[0020] When < C1, it is considered that the potential change probability of the terrain in the current area is relatively low, and there is no need to update the dynamic environment map;

[0021] Initializing the coordinates of the starting node and the target node of the navigation path

[0022] Initializing the RRT tree ;

[0023] Setting rule constraint conditions, generating sampling points based on the rule constraint conditions in the updated dynamic environment map , and based on the candidate nodes in the current path , finding the node closest to the sampling point ;

[0024] Connecting the node closest to the sampling point in the current RRT tree T and the sampling point , and through iterative expansion, generating a navigation path from the starting node to the target node .

[0025] As a preferred solution of the underwater terrain survey method based on unmanned ship data collection according to the present invention, wherein: collecting terrain data based on the navigation path and generating a three-dimensional terrain model, and using U-Net to detect the shadow area in the three-dimensional terrain model, the specific steps are as follows:

[0026] The unmanned ship sequentially visits path nodes along the planned navigation path, and at each path node, initial underwater terrain data is collected through sensor devices; the path nodes include the starting node of the unmanned ship on the navigation path , the node closest to the distance sampling point and the target node ;

[0027] The Delaunay triangulation method is used to convert the initial terrain data into triangular grid data, and the triangular grid data is imported into OpenGL to generate a three-dimensional terrain model;

[0028] The three-dimensional terrain model is projected into a two-dimensional depth map, and the two-dimensional depth map is input into the U-Net network. The encoder in the U-Net network is used to extract terrain features, and the decoder in the U-Net network is used to decode the terrain features to generate a binary shadow mask with the same size as the two-dimensional depth map. Based on the binary shadow mask, the shadow area in the three-dimensional terrain model is identified and marked.

[0029] As a preferred solution of the underwater terrain survey method based on unmanned ship data collection according to the present invention, wherein: according to the detected shadow area, supplementary sampling points are generated through geometric analysis, the navigation path is adjusted according to the supplementary sampling points, the shadow part is supplemented with samples, and supplementary sampling data is obtained. The specific steps are as follows:

[0030] Using geometric analysis methods, the boundary contour of the shadow area is extracted to form a set of boundary points B;

[0031] The shadow area is discretely divided to generate regular grids covering the shadow area;

[0032] Based on the set of boundary points B, the center coordinates of each regular grid are used as supplementary sampling points to generate a set of supplementary sampling points V;

[0033] Calculate the distances between the supplementary sampling points in the set of supplementary sampling points V and the current position of the unmanned ship, and insert the supplementary sampling points with the shortest distances into the current navigation path in sequence. Based on the rule constraint conditions, the current navigation path is adjusted;

[0034] The shadow area is supplemented with samples according to the adjusted navigation path to obtain supplementary sampling data.

[0035] As a preferred solution of the underwater terrain survey method based on unmanned ship data collection according to the present invention, wherein: the supplementary sampling data is fused with the three-dimensional terrain model to generate a terrain survey report. The specific steps are as follows:

[0036] The supplementary sampling point data is fused into the initial terrain data of the three-dimensional terrain model to generate complete gridded terrain data; ​

[0037] Determine whether the grid cell corresponding to the supplementary sampling point is empty;

[0038] If the grid cell corresponding to the supplementary sampling point is empty, directly fill in the supplementary sampling data; integrate the filled supplementary sampling point data into the initial terrain data of the three-dimensional terrain model to generate complete gridded terrain data;

[0039] If the grid cell corresponding to the supplementary sampling point is not empty, directly integrate the supplementary sampling point data into the initial terrain data of the three-dimensional terrain model to generate complete gridded terrain data;

[0040] Update the three-dimensional terrain model based on the gridded terrain data, and generate a terrain survey report using the updated three-dimensional terrain model.

[0041] In a second aspect, the present invention provides an underwater terrain survey system based on unmanned ship data collection, including an environmental map generation module, a terrain change prediction module, a path planning module, a shadow area detection module, a shadow area supplementary sampling module, and a report generation module; Environmental map generation module: used to generate a dynamic environmental map based on the collected multi-modal environmental data; Terrain change prediction module: used to dynamically perceive the underwater environment based on the dynamic environmental map and predict potential terrain changes using the deep Q-learning algorithm; Path planning module: used to determine whether to update the dynamic environmental map based on the predicted potential terrain changes, and plan a navigation path based on the updated dynamic environmental map using the RRT algorithm; Shadow area detection module: used to collect terrain data based on the navigation path and generate a three-dimensional terrain model, and detect shadow areas in the three-dimensional terrain model using U-Net; Shadow area supplementary sampling module: used to generate supplementary sampling points through geometric analysis according to the detected shadow areas, adjust the navigation path according to the supplementary sampling points, supplement the shadow parts, and obtain supplementary sampling data; Report generation module: used to integrate the supplementary sampling data with the three-dimensional terrain model to generate a terrain survey report.

[0042] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, it implements any step of the underwater terrain survey method based on unmanned ship data collection as described in the first aspect of the present invention.

[0043] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, it implements any step of the underwater terrain survey method based on unmanned ship data collection as described in the first aspect of the present invention.

[0044] The beneficial effects of the present invention are as follows: By means of dynamic environment perception and path planning, as well as detecting and supplementing the acquisition of shadow areas, the efficiency and accuracy of underwater terrain survey are improved. The deep Q-learning algorithm is used to achieve dynamic environment perception and path optimization, ensuring that the unmanned ship can efficiently avoid obstacles and cover the target area; the U-Net is used to detect the uncovered areas, combined with geometric analysis to generate supplement acquisition points, and adjust the path to complete precise supplement acquisition, solving the problem of model blind spots. Finally, a complete and high-precision three-dimensional terrain model is generated, overcoming the deficiencies of the prior art in path planning and modeling integrity. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0046] Figure 1 It is a flowchart of the underwater terrain survey method based on unmanned ship data acquisition in Embodiment 1.

[0047] Figure 2 It is a schematic diagram of the underwater terrain survey system based on unmanned ship data acquisition in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0049] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0050] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure or characteristic that can be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude other embodiments.

[0051] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides an underwater terrain survey method based on unmanned ship data acquisition, including the following steps:

[0052] S1: Generate a dynamic environmental map based on the collected multi-modal environmental data, specifically including:

[0053] S1.1: Collect multi-modal environmental data in real time through sensors; the multi-modal environmental data includes underwater terrain, obstacle distribution, and water flow velocity.

[0054] It should be noted that the above data provides basic support for generating a dynamic environmental map, planning a safe navigation path, and accurately surveying the underwater terrain.

[0055] S1.2: Use Bayesian fusion to fuse the multi-modal environmental data, and combine GPS to perform spatial positioning on the multi-modal environmental data, and generate the navigation trajectory of the unmanned ship according to the spatial positioning.

[0056] Furthermore, after preprocessing the multi-modal environmental data collected by each sensor, extracting target features and calibrating data noise, the present invention uses Bayes' formula to fuse these data. By setting the prior probability (such as the obstacle position estimation based on historical data) and the likelihood function (such as the probability distribution of sonar echo intensity and water flow disturbance), the confidence of each modal data is quantified, and the posterior probability is calculated to achieve data fusion, thereby improving the accuracy and consistency of the environmental data. Subsequently, combined with the accurate longitude and latitude information provided by GPS, the fused multi-modal data is mapped into a unified geographic coordinate system to ensure that the unmanned ship can accurately locate itself and surrounding environmental features, such as the exact positions of underwater terrain and obstacles. Based on these high-precision spatial positioning data, the system generates an optimized and safe navigation trajectory to guide the unmanned ship to avoid potential dangers and efficiently complete tasks, such as bypassing newly identified obstacles or adjusting the path to adapt to water flow changes.

[0057] It should be noted that when performing Bayesian fusion of multi-modal environmental data, first preprocess the data collected by each sensor, including denoising, feature extraction, and time synchronization, to ensure the quality and consistency of the data. Next, by setting the prior probability (a preliminary estimate of the environmental state based on existing knowledge or historical data) and the likelihood function (describing the relationship between new observed data and different environmental hypotheses), combine information such as the obstacle position detected by sonar and water flow disturbance. Use Bayes' formula to update these probabilities and calculate the posterior probability, thereby quantifying the confidence of each modal data and achieving effective data fusion. For example, when determining the exact position of an underwater obstacle, the present invention not only considers the potential obstacle position indicated by the sonar echo intensity, but also combines the changes in the water flow velocity field to infer the water flow disturbance that may affect the accuracy of sonar readings, and finally obtains a more credible obstacle positioning result through Bayesian inference. This process significantly improves the accuracy and reliability of environmental perception and provides a solid foundation for subsequent path planning and obstacle avoidance.

[0058] S1.3: Generate a dynamic environment map through Delaunay based on the navigation trajectory of the unmanned ship and multi-modal environmental data.

[0059] Furthermore, based on the real-time navigation trajectory of the unmanned ship and multi-modal environmental data, the present invention dynamically generates and continuously updates the environment map through the Delaunay triangulation method to accurately reflect features such as underwater terrain, obstacle distribution, and water flow velocity. When the unmanned ship sails along the planned or adaptively adjusted path, it continuously collects discrete point data using sensors (such as sonar, visual cameras, water flow velocity meters), and this data includes information such as terrain depth, obstacle boundaries, and water flow velocity. These multi-modal data are combined with the current position of the unmanned ship (provided by GPS) to ensure that each measurement point can be accurately mapped into the geographical coordinate system. Then, the Delaunay triangulation algorithm is applied to transform these discrete points into an optimal and non-overlapping triangular grid to construct a continuous three-dimensional environment model. As the unmanned ship continues to move along its trajectory, new multi-modal data are continuously collected and incorporated into the existing map, enabling the environment map to be updated in real time and dynamically reflect the changes in the current underwater environment. This method of combining the navigation trajectory of the unmanned ship and multi-modal environmental data not only improves the accuracy and consistency of environmental perception but also provides reliable obstacle avoidance and path optimization support for the unmanned ship to ensure its safe and efficient task execution in complex environments.

[0060] S2: Based on the dynamic environment map, use the deep Q-learning algorithm to dynamically perceive the underwater environment and predict the probability of potential terrain changes, specifically including:

[0061] S2.1: Through feature extraction and discretization methods, transform the data features in the dynamic environment map into the state space of the deep Q-learning algorithm, and use the data features of the local area of the current underwater environment as the state in the state space. The data features include terrain depth, obstacle position, water flow intensity, etc.

[0062] S2.2: Define the action space by taking the operations of the unmanned ship in the current environment as actions.

[0063] Furthermore, the actions in the action space of the unmanned ship can include a series of operations related to navigation and task execution, such as moving forward, backward, turning left, turning right, staying, and changing the navigation speed, etc. These actions are refined according to the physical characteristics of the unmanned ship (such as turning radius, maximum speed, acceleration, etc.) and environmental constraints (such as water flow direction, obstacle position, and terrain features). For example, in a complex underwater environment, the action space can also be extended to specific operations such as "bypassing obstacles" or "adjusting depth" to adapt to task requirements. In addition, each action corresponds to different navigation results (such as path length, energy consumption, survey coverage rate, etc.)

[0064] S2.3: Calculate the comprehensive reward value based on the distance between the unmanned ship and the obstacle, the target area coverage ratio, and the navigation cost. The specific expression is:

[0065]

[0066] where, represents the distance between the unmanned ship and the obstacle when performing action in state ; represents the target area coverage ratio surveyed when performing action in state ; represents the navigation cost required when performing action in state ; represents the comprehensive reward value when the unmanned ship performs action a in state s.

[0067] It should be noted that is to calculate the closest distance between the unmanned ship and the obstacle (calculated by Euclidean distance), is obtained by calculating the grid coverage rate of the current path, is calculated according to the path length and energy consumption.

[0068] The present invention detects the distance between the unmanned ship and the obstacle in real time through a sensor, and calculates the closest distance using the Euclidean distance to evaluate the safety of obstacle avoidance.

[0069] The present invention calculates the target area coverage ratio when the unmanned ship performs action in state according to the ratio of the number of grid cells covered by the current path of the unmanned ship to the total number of grid cells in the target area .

[0070] The present invention calculates the navigation cost required for the unmanned ship to complete the action in state according to the path length, speed, and energy consumption model when the unmanned ship performs action .

[0071] S2.4: Based on the comprehensive reward value , combined with the divergence degree of the water flow velocity field, terrain depth, and obstacle density in the underwater environment, predict the potential terrain change probability. The expression is:

[0072]

[0073] where, represents at the coordinate position and time The predicted potential probability of terrain change Indicates the divergence degree of the water flow at the coordinate position and time underneath, Indicates the coordinate position and time of the terrain depth, is the attenuation coefficient of the terrain depth to the water flow disturbance, Indicates at the coordinate position and time underneath, the obstacle density, Indicates the weight coefficient of the influence of obstacles on terrain change, Indicates the coordinate position and time of the water flow velocity.

[0074] It should be noted that the divergence degree of the water flow at the coordinate position and time underneath , is calculated through the divergence of the water flow velocity field, and the expression is:

[0075]

[0076] Among them, is the component of the water flow velocity field in the direction, is the component of the water flow velocity field in the direction, is the partial derivative operator.

[0077] S3: Judge whether to update the dynamic environment map according to the predicted potential probability of terrain change, and plan the navigation path based on the updated dynamic environment map using the RRT algorithm, specifically including:

[0078] S3.1: Define the threshold terrain change risk threshold C1 based on the historical terrain change frequency and change amplitude.

[0079] S3.2: When ≥C1, it is considered that the potential change probability of the terrain in the current area is relatively high, and the dynamic environment map is updated.

[0080] S3.3: When <C1, it is considered that the potential change probability of the terrain in the current area is relatively low, and there is no need to update the dynamic environment map.

[0081] It should be noted that the process of updating the dynamic environment map is to collect the terrain depth, water flow velocity and obstacle data in the area in real time, and fuse or replace the new data with the original map. For example, when the potential change probability of the terrain in a certain area When exceeding the risk threshold C1, re - collect the terrain depth through sonar (e.g., from 10 m to 9.5 m), update the water flow speed by the water flow sensor (e.g., from 1.2 m / s to 1.5 m / s), and detect newly added obstacles (e.g., at position (x + 10, y - 5)). Replace or weight - fuse these data into the corresponding grid to generate an updated dynamic environment map to reflect the latest underwater environment characteristics.

[0082] S3.4: Initialize the starting node of the navigation path and the coordinates of the target node .

[0083] S3.5: Initialize the RRT tree .

[0084] S3.6: Set rule - constraint conditions, generate sampling points based on the rule - constraint conditions in the updated dynamic environment map , and based on the current RRT tree among the candidate nodes , find the node closest to the sampling point , with the expression:

[0085]

[0086] where, is the node in the current RRT tree T closest to the sampling point , and is a candidate node in the tree T.

[0087] It should be noted that setting the rule - constraint conditions specifically means: integrating environmental information such as terrain depth, water flow speed, and obstacle distribution to provide constraints for the navigation path planning of the unmanned ship. For example, when the map shows that the water flow speed in a certain area exceeds 2 m / s (strong - current area), set this area as a "non - passable" restricted area; if the terrain depth in a certain area is less than 3 m (shallow - water area), mark it as a "low - priority" area; while the area with dense obstacles is set as an "obstacle - avoidance area", and the path of the unmanned ship needs to maintain a safety distance of at least 5 m. These rule - constraint conditions will limit the generation of sampling points and guide the unmanned ship to choose a safe and low - cost path for navigation and survey.

[0088] S3.7: Connect the node closest to the sampling point in the current RRT tree T and the sampling point , and through iterative expansion, generate a navigation path from the starting node to the target node .

[0089] ​​​​Specifically: Calculate the sampling points according to the distance formula (such as Euclidean distance). Calculate the distances between the sampling points and all nodes in the tree, and select the node with the minimum distance as the connection starting point. Then, generate an incremental path (expanding by a fixed step size) from in the direction, and at the same time check whether the path meets the rule constraints in the dynamic environment map (such as obstacle avoidance areas, water flow effects, etc.). If the path is valid, add the newly expanded node to the tree T and repeat the above process to gradually expand the scope of the tree until the path connects the starting node and the target node , and finally generate an optimal navigation path that meets all constraint conditions.

[0090] S4: Collect terrain data based on the navigation path and generate a three-dimensional terrain model, and use U-Net to detect the shadow areas in the three-dimensional terrain model, specifically including:

[0091] S4.1: The unmanned ship sequentially visits the path nodes along the planned navigation path, and at each path node, collects the initial underwater terrain data through sensor devices; the path nodes include the starting node of the unmanned ship on the navigation path , the node closest to the distance sampling point , and the target node . .

[0092] S4.2: Use the Delaunay triangulation method to convert the initial terrain data into triangular mesh data, and import the triangular mesh data into OpenGL to generate a three-dimensional terrain model.

[0093] S4.3: Project the three-dimensional terrain model into a two-dimensional depth map, input the two-dimensional depth map into the U-Net network, use the encoder in the U-Net network to extract terrain features, and use the decoder in the U-Net network to decode the terrain features to generate a binary shadow mask with the same size as the two-dimensional depth map, and identify and mark the shadow areas in the three-dimensional terrain model based on the binary shadow mask.

[0094] Specifically: The three-dimensional terrain model is converted into a two-dimensional depth map through orthographic projection or perspective projection. The generated two-dimensional image retains the depth information of the terrain (such as terrain undulations and height changes) and is passed as input to the U-Net network. The encoder part of the U-Net extracts terrain features in the topographic map through multi-layer convolution and pooling operations, such as the boundary and shape characteristics of the shadow area; subsequently, the decoder in the U-Net gradually restores the image resolution through deconvolution and upsampling operations and generates a binary shadow mask with the same size as the two-dimensional depth map. Based on the generated binary shadow mask, the identified shadow area is remapped back to the three-dimensional terrain model to accurately mark the shadow area in the three-dimensional terrain model for subsequent data supplement collection or terrain analysis, where "1" represents the shadow area and "0" represents the non-shadow area.

[0095] S5: According to the detected shadow area, generate supplement collection points through geometric analysis, adjust the navigation path according to the supplement collection points, and perform supplement collection on the shadow part to obtain supplement collection data, specifically including:

[0096] S5.1: Using geometric analysis methods, extract the boundary contour of the shadow area to form a boundary point set B.

[0097] For example: Through morphological operations (such as edge detection or convex hull algorithm), extract the boundary of the shadow area to ensure that the supplement collection points cover the edge of the shadow area.

[0098] S5.2: Based on the boundary point set B, take the center coordinates of each regular grid as supplement collection points to generate a supplement collection point set V. Among them, the regular grid is obtained by discretely dividing the shadow area.

[0099] S5.3: Calculate the distance between each supplement collection point in the supplement collection point set V and the current position of the unmanned ship, and insert the supplement collection point with the closest distance into the current navigation path in sequence. Based on the rule constraint conditions, adjust the current navigation path.

[0100] Furthermore, according to the supplement collection point set V, first calculate the distance between each supplement collection point and the current position of the unmanned ship (such as through the Euclidean distance formula), and sort the supplement collection points from near to far according to the distance. Then, insert the closest supplement collection point into the current navigation path in sequence to ensure that the unmanned ship preferentially visits the supplement collection points with closer distances, thereby optimizing the navigation efficiency. When inserting supplement collection points, refer to the rule constraint conditions in the dynamic environment map (such as obstacle avoidance areas, shallow water areas, or strong current areas) to adjust the path to avoid high-risk areas.

[0101] For example, if the unmanned ship is currently located at (x0, y0) and the set of supplementary sampling points is {(x1, y1), (x2, y2), (x3, y3)}, after calculating the distances and sorting, it is (x2, y2) < (x1, y1) < (x3, y3), and the unmanned ship first navigates to (x2, y2). If the path from (x0, y0) to (x2, y2) passes through the obstacle area, the path planning algorithm (such as A* or RRT) will adjust the navigation route based on the dynamic environment map, bypass the obstacles and reach the supplementary sampling point to ensure safety and navigation efficiency.

[0102] S5.4: Perform supplementary sampling on the shaded area according to the adjusted navigation path to obtain supplementary sampling data.

[0103] It should be noted that performing supplementary sampling on the shaded area according to the adjusted navigation path can efficiently cover the shaded areas that were not surveyed due to complex terrain or sensor blind spots. By optimizing the path, the navigation time and energy consumption can be reduced, while ensuring the integrity and accuracy of the supplementary sampling data. This not only significantly improves the integrity and resolution of the three-dimensional terrain model, but also provides more reliable data support for subsequent terrain analysis, environmental assessment, and navigation decision-making, avoiding potential risks or misjudgments caused by data gaps.

[0104] S6: Integrate the supplementary sampling data with the three-dimensional terrain model to generate a terrain survey report, which specifically includes:

[0105] S6.1: Integrate the supplementary sampling point data into the initial terrain data of the three-dimensional terrain model to generate complete gridded terrain data.

[0106] S6.2: Determine whether the grid cell corresponding to the supplementary sampling point is empty.

[0107] S6.2.1: If the grid cell corresponding to the supplementary sampling point is empty, directly fill in the supplementary sampling data; integrate the filled supplementary sampling point data into the initial terrain data of the three-dimensional terrain model to generate complete gridded terrain data.

[0108] For example, when the unmanned ship obtains new depth data during supplementary sampling in a certain shaded area, if there is no data in the grid cell at this location, directly fill in the newly collected data into this empty cell.

[0109] S6.2.2: If the grid cell corresponding to the supplementary sampling point is not empty, directly integrate the supplementary sampling point data into the initial terrain data of the three-dimensional terrain model to generate complete gridded terrain data.

[0110] For example, when there is old data at the supplementary sampling point, average the new and old data or integrate them according to a certain algorithm to update the data of this grid cell and ensure the accuracy of the terrain model.

[0111] S6.2.3: Update the 3D terrain model based on the gridded terrain data, and generate a terrain survey report using the updated 3D terrain model.

[0112] Furthermore, after all supplementary survey data has been properly incorporated into the 3D terrain model, complex algorithms will be executed to recalculate and optimize terrain features, ensuring that each grid cell accurately reflects the latest survey results, and conducting a detailed analysis of slope, depth changes, and potential geological structures. Based on the updated high-precision 3D terrain model, a detailed underwater terrain survey report will be automatically generated. This report not only summarizes the terrain changes, specifically marks the locations of new highlands, depressions, and obstacles that may affect navigation safety, but also provides a comparative analysis with the previous terrain data to help users understand the terrain change trends and their potential impacts. Additionally, it includes a series of intuitive charts and images to assist professionals in in-depth research and decision-making.

[0113] This embodiment also provides an underwater terrain survey system based on unmanned ship data collection, including: an environmental map generation module, a terrain change prediction module, a path planning module, a shadow area detection module, a shadow area supplementary survey module, and a report generation module; The environmental map generation module: is used to generate a dynamic environmental map based on the collected multi-modal environmental data; The terrain change prediction module: is used to dynamically perceive the underwater environment based on the dynamic environmental map using the deep Q-learning algorithm and predict potential terrain changes; The path planning module: is used to determine whether to update the dynamic environmental map based on the predicted potential terrain changes, and plan a navigation path based on the updated dynamic environmental map using the RRT algorithm; The shadow area detection module: is used to collect terrain data based on the navigation path and generate a 3D terrain model, and detect shadow areas in the 3D terrain model using U-Net; The shadow area supplementary survey module: is used to generate supplementary survey points through geometric analysis according to the detected shadow areas, adjust the navigation path according to the supplementary survey points, conduct supplementary survey on the shadow parts, and obtain supplementary survey data; The report generation module: is used to fuse the supplementary survey data with the 3D terrain model to generate a terrain survey report.

[0114] This embodiment also provides a computer device applicable to the case of the underwater terrain survey method based on unmanned ship data collection, including: a memory and a processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the underwater terrain survey method based on unmanned ship data collection as proposed in the above embodiment.

[0115] The computer device can be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads provided on the outer shell of the computer device. It can also be an external keyboard, touchpad, or mouse, etc.

[0116] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the underwater terrain survey method based on unmanned ship data acquisition proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0117] In summary, the present invention improves the efficiency and accuracy of underwater terrain survey through: dynamic environment perception and path planning, and shadow area detection and supplementary acquisition. The depth Q-learning algorithm is used to achieve dynamic environment perception and path optimization, ensuring that the unmanned ship can efficiently avoid obstacles and cover the target area. The U-Net is used to detect uncovered areas, combined with geometric analysis to generate supplementary acquisition points, and the path is adjusted to complete precise supplementary acquisition, solving the problem of model blind spots. Finally, a complete and high-precision three-dimensional terrain model is generated, overcoming the deficiencies of the prior art in path planning and modeling integrity.

[0118] Example 2. Referring to Table 1, this is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of the underwater terrain survey method based on unmanned ship data collection are given.

[0119] To verify the advantages of the underwater terrain survey method based on unmanned ship data collection, an underwater terrain survey experiment was designed. The experimental area was selected in a coastal water area of 500m×500m, with a water depth range of 3.2m to 14.8m. There were natural obstacles (such as reefs, shoals) and complex water flow disturbances in the area. The unmanned ship was equipped with multi-modal sensors (multi-beam sonar, current meter and GPS) to collect underwater environment data, and completed terrain survey and supplementary collection of shadow areas through dynamic perception algorithms and path planning methods.

[0120] First, the unmanned ship uses multi-modal sensors (such as multi-beam sonar, current meter and GPS) to collect data such as underwater terrain depth, water flow velocity and obstacle distribution in real time in the test area, and uses the Bayesian fusion algorithm to process the multi-modal data, generates a preliminary environmental map by combining Delaunay triangulation, and dynamically updates the environmental map according to the navigation trajectory of the unmanned ship, providing a basis for subsequent terrain change prediction and path planning.

[0121] Secondly, based on the dynamic environmental map, the characteristics of the underwater environment are perceived through the deep Q-learning algorithm, and the predicted probability of potential terrain changes is marked in real time. The RRT algorithm is combined to optimize the navigation path, ensuring that the unmanned ship avoids high-risk areas and covers the target survey area as much as possible, thereby improving the safety and efficiency of path planning.

[0122] Then, the unmanned ship visits each node in turn according to the planned navigation path, collects the initial terrain data, and generates a preliminary three-dimensional terrain model through Delauna triangulation. Subsequently, the three-dimensional terrain model is projected into a two-dimensional depth map, the shadow area in the U-Net network detection model is input, supplementary collection points are generated through geometric analysis, and the navigation path is adjusted according to the dynamic environmental map to supplement the shadow area and update the missing data.

[0123] Finally, the supplementary collection data is fused with the initial terrain data to generate complete grid terrain data, and the three-dimensional terrain model is updated. Analyze the key data indicators of the model before and after supplementary collection, including terrain depth accuracy, shadow area ratio, path coverage rate, navigation path length and energy consumption, etc. Finally, it is concluded that the terrain model after supplementary collection has a significant improvement in coverage rate, accuracy and efficiency, verifying the superiority of the method of the present invention.

[0124] In this experiment, the prior art uses traditional underwater terrain survey methods. It mainly collects underwater terrain data through a multibeam sonar carried by an unmanned boat and sequentially covers the target area according to a fixed navigation path. In the initially generated three-dimensional terrain model, data collection only relies on the preset path and fails to dynamically sense environmental changes, resulting in problems such as data gaps or reduced accuracy in complex terrain and high-flow areas (such as shoals and areas with dense obstacles). In addition, the prior art lacks an effective shadow area detection and supplementary collection mechanism, leading to a relatively high proportion of shadow areas and making it difficult for the integrity and accuracy of the three-dimensional terrain model to meet the requirements of accurate surveying.

[0125] Specifically, it is shown in Table 1 below:

[0126] Table 1 Comparison Table of Underwater Terrain Survey Test Data

[0127] Parameter Name Prior Art The Present Invention Enhancement Rate (%) Unit Terrain Depth Accuracy (Average Error) 3.87 1.23 68.2 meter Ratio of Shadow Area 17.46 2.85 83.7 % Path Coverage Rate 74.31 96.78 30.2 % Navigation Path Length 5274.56 4928.34 -6.6 meter Total Energy Consumption 1215.43 1158.67 -4.7 kilowatt-hour

[0128] Through the data analysis of the above table, it can be clearly seen that the present invention has significant advantages over the prior art in terms of the accuracy, coverage rate, and efficiency of underwater terrain survey. The present invention realizes high-precision and high-coverage survey of complex terrain through innovative technologies such as dynamic environment perception, path optimization, and shadow area supplementary collection, while reducing the navigation cost and energy consumption. For example, in terms of the terrain depth accuracy, the average error of the prior art is 3.87 meters, while the present invention significantly reduces the error to 1.23 meters through precise shadow area supplementary collection and dynamic path adjustment, with the accuracy improved by 68.2%. The proportion of the shadow area also decreases significantly from 17.46% of the prior art to 2.85%, and the coverage rate is increased by 83.7%, indicating that the present invention can effectively identify and supplement the blind areas that are difficult to cover by traditional methods. This benefits from the accurate shadow detection of the U-Net network and the design of supplementary collection points generated by geometric analysis.

[0129] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An underwater terrain survey method based on unmanned ship data collection, characterized in that: include, Generate dynamic environment maps based on collected multimodal environment data; Based on the dynamic environment map, the deep Q-learning algorithm is used to dynamically perceive the underwater environment and predict the probability of potential terrain changes; Determine whether to update the dynamic environment map based on the predicted probability of potential terrain changes, and plan the navigation path using the updated dynamic environment map based on the RRT algorithm; Collect terrain data based on the navigation path and generate a 3D terrain model, and use U-Net to detect shadow areas in the 3D terrain model; According to the detected shadow area, the supplementary mining points are generated through geometric analysis, and the navigation path is adjusted according to the supplementary mining points to supplement the shadow area and obtain supplementary mining data; The supplementary mining data is integrated with the three-dimensional terrain model to generate a terrain survey report.

2. The underwater topographic survey method based on unmanned ship data collection according to claim 1, characterized in that: The specific steps of generating a dynamic environment map based on the collected multimodal environment data are as follows: Collect multimodal environmental data in real time through sensors; multimodal environmental data includes underwater terrain, obstacle distribution and water flow speed; The multimodal environmental data are fused by using Bayesian fusion, and the multimodal environmental data are spatially positioned in combination with GPS, and the navigation trajectory of the unmanned ship is generated according to the spatial positioning; Based on the navigation trajectory of the unmanned ship and multimodal environmental data, a dynamic environmental map is generated through Delaunay.

3. The underwater topographic survey method based on unmanned ship data collection according to claim 2, characterized in that: Based on the dynamic environment map, the deep Q learning algorithm is used to dynamically perceive the underwater environment and predict the probability of potential terrain changes. The specific steps are as follows: Through feature extraction and discretization methods, the data features in the dynamic environment map are converted into the state space of the deep Q learning algorithm, and the data features of the local area of ​​the current underwater environment are used as the state in the state space; The operation of the unmanned ship in the current environment is taken as an action and the action space is defined; The comprehensive reward value is calculated based on the distance between the unmanned ship and the obstacle, the coverage ratio of the target area and the navigation cost. ; Based on comprehensive reward value , combining the divergence of the water velocity field in the underwater environment, the depth of the terrain, and the density of obstacles to predict the probability of potential terrain changes .

4. The underwater topographic survey method based on unmanned ship data collection according to claim 3, characterized in that: The specific steps of determining whether to update the dynamic environment map according to the predicted probability of potential terrain changes, and planning the navigation path using the updated dynamic environment map based on the RRT algorithm are as follows: Based on the frequency and magnitude of historical terrain changes, the terrain change risk threshold C1 is defined; when When ≥C1, it is considered that the potential change probability of the terrain in the current area is high, and the dynamic environment map is updated; when When <C1, it is considered that the potential change probability of the current regional terrain is low, and there is no need to update the dynamic environment map; Initialize the starting node of the navigation path and the target node Coordinates Initialize the RRT tree ; Set rule constraints and generate sampling points based on the rule constraints in the updated dynamic environment map , and based on the candidate nodes in the current path , find the sampling point The closest node ; Connect the distance sampling points in the current RRT tree T The closest node and sampling points , and through iterative expansion, generate To the target node navigation path.

5. The underwater topographic survey method based on unmanned ship data collection according to claim 4, characterized in that: The steps of collecting terrain data based on the navigation path and generating a three-dimensional terrain model and detecting the shadow area in the three-dimensional terrain model using U-Net are as follows: The unmanned ship visits the path nodes in sequence along the planned navigation path, and collects the initial underwater terrain data through the sensor equipment at each path node; the path nodes include the starting node of the unmanned ship on the navigation path , distance sampling point The closest node and the target node ; Use Delaunay triangulation method to convert initial terrain data into triangular mesh data, and import the triangular mesh data into OpenGL to generate a three-dimensional terrain model; The 3D terrain model is projected into a 2D depth map, and the 2D depth map is input into the U-Net network. The terrain features are extracted by the encoder in the U-Net network, and the terrain features are decoded by the decoder in the U-Net network to generate a binary shadow mask with the same size as the 2D depth map. The shadow area in the 3D terrain model is identified and marked based on the binary shadow mask.

6. The underwater topographic survey method based on unmanned ship data collection according to claim 5, characterized in that: The method generates supplementary mining points through geometric analysis based on the detected shadow area, adjusts the navigation path according to the supplementary mining points, supplements the mining of the shadow area, and obtains supplementary mining data. The specific steps are as follows: Using geometric analysis methods, the boundary contour of the shadow area is extracted to form a boundary point set B; Discretize the shadow area and generate a regular grid covering the shadow area; Based on the boundary point set B, the center coordinates of each regular grid are used as supplementary mining points to generate a supplementary mining point set V; Calculate the distance between each supplementary mining point in the supplementary mining point set V and the current position of the unmanned ship, and insert the nearest supplementary mining points into the current navigation path in sequence, and adjust the current navigation path based on the rule constraints; The shadow area is supplemented with samples according to the adjusted navigation path to obtain supplementary sampling data.

7. The underwater topographic survey method based on unmanned ship data collection according to claim 6, characterized in that: The above-mentioned steps of fusing the supplementary mining data with the three-dimensional terrain model to generate a terrain survey report are as follows: The supplementary mining point data is integrated into the initial terrain data of the 3D terrain model to generate complete gridded terrain data; Determine whether the grid cell corresponding to the supplementary mining point is empty; If the grid cell corresponding to the supplementary mining point is empty, the supplementary mining data is directly filled in; the filled supplementary mining point data is merged into the initial terrain data of the three-dimensional terrain model to generate complete gridded terrain data; If the grid cell corresponding to the supplementary mining point is not empty, the supplementary mining point data is directly fused into the initial terrain data of the three-dimensional terrain model to generate complete gridded terrain data; The three-dimensional terrain model is updated based on the gridded terrain data, and a terrain survey report is generated using the updated three-dimensional terrain model.

8. An underwater terrain survey system based on unmanned ship data collection, based on the underwater terrain survey method based on unmanned ship data collection according to any one of claims 1 to 7, characterized in that: It includes an environment map generation module, a terrain change prediction module, a path planning module, a shadow area detection module, a shadow area supplementary sampling module and a report generation module; Environmental map generation module: used to generate dynamic environmental maps based on the collected multimodal environmental data; Terrain change prediction module: used to dynamically perceive the underwater environment based on the dynamic environment map and predict potential terrain changes using the deep Q learning algorithm; Path planning module: used to determine whether to update the dynamic environment map based on the predicted potential terrain changes, and plan the navigation path using the updated dynamic environment map based on the RRT algorithm; Shadow area detection module: used to collect terrain data based on the navigation path and generate a 3D terrain model, and use U-Net to detect shadow areas in the 3D terrain model; Shadow area supplementary mining module: It is used to generate supplementary mining points through geometric analysis according to the detected shadow area, adjust the navigation path according to the supplementary mining points, supplementary mine the shadow area, and obtain supplementary mining data; Report generation module: used to integrate the supplementary mining data with the three-dimensional terrain model to generate a terrain survey report.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the underwater terrain survey method based on unmanned ship data collection according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the underwater terrain survey method based on unmanned ship data collection according to any one of claims 1 to 7 are implemented.

Images