Water network high-resolution terrain scanning path planning method based on multi-source information

Abstract

The invention relates to the technical field of water path planning, in particular to a water network high-resolution terrain scanning path planning method based on multi-source information. The method comprises the following steps: obtaining water surface height data through a water level station along a water network, fusing multi-source low-resolution historical geographic space data, and constructing a water body thickness basic data set; and performing mirroring processing on the water body thickness basic data set in the vertical direction, constructing a water body physical model and an upper surface irregular triangulation network, realizing beam footprint position reduction of the multi-beam sounding system, designing an optimal multi-beam scanning path, and realizing high-resolution full-coverage efficient observation of a water network. According to the method, a full-flow path generation and feedback optimization mechanism driven by multi-source hydrological-topographic data is constructed, so that the problems of serious path overlapping, high regional blind scanning rate and lack of topographic response capability in traditional underwater scanning path planning are solved, and high-resolution observation of water network spatial information and thorough perception of underwater work conditions are realized.

Description

Technical Field

[0001] This invention relates to the field of water path planning technology, and in particular to a method for high-resolution topographic surveying and path planning of water networks based on multi-source information. Background Technology

[0002] Existing water network topographic survey path planning technologies mostly rely on single measurement methods, such as riverbed DEMs (Digital Elevation Models) generated from traditional single-beam or multi-beam bathymetry data. This lack of effective integration and utilization of multi-source heterogeneous information leads to frequent problems such as slow response to topographic changes, high scanning redundancy, and insufficient coverage of low-lying areas. On the one hand, existing path planning methods often employ regular rasterized paths or fixed-interval parallel scanning strategies, failing to dynamically adjust according to spatial heterogeneity such as underwater topographic slope, channel depth changes, and river bends. This easily results in paths that do not match the actual terrain, leading to low scanning efficiency or coverage blind spots. On the other hand, when processing hydrological information such as upstream and downstream water level changes and dynamic evolution of water thickness, existing technologies often only use it as post-processing reference data, failing to fully integrate it into the path generation stage, thus lacking real-time coupling judgment of water level and topographic interaction changes. Furthermore, current methods lack a closed-loop feedback mechanism for optimizing and controlling multi-temporal scanning paths. They fail to design path reconstruction or supplementary scanning strategies based on historical scanning data or hole area identification results, resulting in a persistently high data hole rate and making it difficult to guarantee the accuracy of subsequent analysis and terrain reconstruction. Summary of the Invention

[0003] Therefore, it is necessary to provide a high-resolution topographic survey path planning method for water networks based on multi-source information to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a high-resolution topographic survey path planning method for water networks based on multi-source information is provided, the method comprising the following steps: Step S1: Collect multi-source historical data of the water network, integrate water level data from water level stations, and construct basic data on water body thickness; Step S2: Perform vertical mirroring on the basic dataset of water body thickness and construct a vertical mirroring algorithm to output a water body physical model; construct an irregular triangular network based on the water body physical model and generate a simulated underwater terrain morphology; divide the simulated underwater terrain morphology into sub-regions and calculate the distance between adjacent paths to obtain the maximum distance data of underground paths. Step S3: Extract the river centerline based on the simulated underwater terrain morphology, and perform density partitioning according to the preset slope angle threshold to obtain waterway slope density partitioning; generate terrain adaptive paths for waterway slope density partitioning, and perform global overlap rate verification to obtain waterway adaptive scanning paths. Step S4: Perform a full-process path evaluation on the waterway adaptive scanning path to obtain multi-source water network scanning path evaluation data; based on the multi-source water network scanning path evaluation data, perform parameter feedback on the adjacent path spacing calculation formula to obtain the waterway planning feedback parameter set; based on the multi-source water network scanning path evaluation data and the waterway planning feedback parameter set, perform high-resolution, full-coverage scanning path planning for the water network multibeam echo sounding system.

[0005] The beneficial effects of this invention are that it enables continuous monitoring of upstream and downstream water levels through the deployment of water level stations. After collecting multi-source water level data, it synchronizes historical depth data and generates a basic dataset of water body thickness. This dataset has both temporal consistency and spatial correspondence, providing accurate boundary conditions for subsequent inversion modeling. In step S2, the water body thickness data is mirrored in the vertical direction of the terrain, and an irregular triangular network (TIN) is constructed in combination with elevation differences. This accurately reproduces the spatial undulations of the underwater landform, especially showing good terrain restoration capabilities in irregular river areas. Based on this, by dividing the TIN model into sub-regions and calculating path spacing, the maximum spacing data between the scanning blind zone and path coverage is effectively identified, thus providing a quantitative basis for the construction of subsequent path distribution strategies. Step S3 further divides the river centerline extracted from the triangular network model into slope angle partitions, dividing areas with different degrees of terrain change. Based on density characteristics, terrain-adaptive paths are automatically generated, and a global overlap rate verification mechanism is introduced to optimize scanning uniformity and coverage integrity from the perspective of path spatial layout. Step S4 introduces a path evaluation model to diagnose the performance of all scanning paths and feeds back evaluation parameters such as overlap rate, void rate, and path efficiency to guide the iterative path adjustment process and achieve adaptive optimization of path planning. Furthermore, a high-resolution, full-coverage scanning path planning report is automatically generated based on the multi-source hydrological-topographic data, visually presenting the path-topography-coverage relationship and providing a quantitative decision-making basis for underwater operations. Therefore, this invention, by constructing a multi-source hydrological-topographic data-driven end-to-end path generation and feedback optimization mechanism, solves the problems of severe path overlap, high blind scan rate, and lack of terrain response capability in traditional underwater scanning path planning, improving the scientific rigor, accuracy, and practicality of underwater operation path planning in multi-source hydrological network areas.

[0006] Preferably, step S1 includes the following steps: Step S11: Obtain the distance between upstream and downstream water level stations; deploy water level stations upstream and downstream of the target waterway, use pressure-type water level gauges to continuously observe the water level at a sampling frequency of 1Hz, and collect water level data from the water level stations. Step S12: Collect multi-source historical data of the water network, and collect single-beam bathymetry historical data from the multi-source historical data of the water network. Use the cross-section method to obtain the elevation of the underwater measuring point. Step S13: Grid the historical single-beam bathymetry data, generate interpolation data, and generate riverbed DEM data; Step S14: Synchronize the water level data from the water level stations with time, and use linear interpolation to calculate the water level data for the entire water area, taking into account the distance between upstream and downstream water level stations. Step S15: Spatially align the riverbed DEM data with the water surface elevation data of the entire water area to generate a basic dataset of water thickness.

[0007] This invention, by acquiring the distance between upstream and downstream water level stations and deploying pressure-type water level gauges with a sampling frequency of 1Hz, can form a high-timeliness, high-resolution water level observation data stream, significantly enhancing the continuity and dynamic change capture capability of hydrological data. Step S12 collects the latest single-beam bathymetry historical data of the waterway and extracts the elevation of the measuring points using the cross-sectional method, obtaining a longitudinal elevation point series with strong terrain reconstruction accuracy, further enhancing the perception accuracy of riverbed morphology. In step S13, the historical bathymetry points are gridded to construct an equally spaced grid structure, and combined with spatial interpolation algorithms (such as IDW or Kriging) to generate a digital elevation model (DEM), forming a bottom terrain representation with a regular structure. Step S14 then performs time synchronization processing on the water level data, and based on the water surface response gradient between the distance between upstream and downstream water level stations and the measurement time, applies a linear interpolation method to reconstruct the water surface elevation data of the entire water area in the spatial dimension, realizing continuous modeling of water level information in spatial distribution. Finally, by unifying and aligning the water surface elevation data with the riverbed DEM data spatially, the water thickness value of each grid cell at the corresponding time can be accurately calculated, forming a basic dataset of water thickness that is both temporally continuous and spatially complete. This dataset not only achieves a complete representation of the three-dimensional structure of the water profile, but also provides a unified physical quantity input interface for various models such as vertical mirror processing terrain modeling, slope zoning analysis, and path density control.

[0008] Preferably, in step S2, the basic dataset of water body thickness is mirrored vertically, and the formula for vertical mirroring is constructed as follows: The basic dataset of water body thickness is mirrored in the vertical direction to obtain mirrored data of water body thickness in the vertical direction. A vertical mirror processing algorithm is constructed based on the mirror processing data of the water body thickness in the vertical direction to obtain a physical model of the water body. The formula for vertical mirror processing is as follows: in, Let be the planar coordinates of any point in the waterway; For point Riverbed elevation data at the location; For point Water surface elevation data at the location; This represents the real-time water level at upstream water level station A. This is the output physical model of the water body.

[0009] This invention utilizes a vertical mirroring formula. Achieving intelligent transformation of three-dimensional coordinate systems, among which (Riverbed elevation) and Difference operation of (water surface elevation) The original underwater concave topography is mapped to convex positive topography, while the baseline term... By establishing a stable reference system through the absolute elevation of the upstream water level, this operation achieves three core gains at the data level: First, it unifies the spatial benchmark, eliminating dynamic fluctuations in water surface elevation (such as those caused by tides and flood discharge). The impact of time-varying disturbances on topographic representation was reduced from ±0.35m to ±0.03m in the Yangtze River case; secondly, topographic features were preserved without loss: mathematical verification showed that the original riverbed slope angle was maintained. Mirror image processing model slope angle satisfy m (verified via the Jacobian matrix) ensures complete inheritance of gradients for key features such as slope inflection and deep pits; thirdly, the algorithm's compatibility is greatly improved: a vertical mirror processing model. The convex structure (ConvexHull ratio > 99%) allows for the direct application of mature terrain eye-following algorithms (such as E-spline path optimization), reducing computational power by 75% compared to traditional underwater path planning (path generation time in the Yangtze River case decreased from 53s to 12s). Data validation shows that this processing improves the slope angle calculation in subsequent dynamic sweep width calculations. The extraction accuracy was improved to 98.7%, and it passed the water level benchmark item 2. Maintain rigid connection with the geodetic coordinate system (translation matrix is...) Ultimately, this technology reduces the length of complex waterway survey paths by 45.1 km (an 89.8% increase in efficiency), while ensuring that the resolution non-uniformity (the ratio of deep to shallow water resolution) is optimized from 4.7 to 1.3, providing millimeter-level topographic base data for digital twins of water networks.

[0010] Preferably, step S2, which involves dividing the simulated underwater terrain into sub-regions and calculating the distance between adjacent paths, includes: The simulated underwater topography is divided into sub-regions according to the contour lines, and the average slope angle of the sub-regions is calculated to obtain the underwater multibeam bathymetry opening angle data. Dynamic sweep width is calculated based on underwater multibeam bathymetry opening angle data to obtain dynamic sweep width interval data for sub-regions. The formula for calculating dynamic sweep width is as follows: Among them, For the i-th sub-region, the sub-region dynamic sweep width interval data; Let be the average slope angle of the i-th sub-region; The angle of the acoustic wave fan-shaped coverage; Let be the water thickness of the i-th sub-region; The distance between adjacent paths is calculated based on the dynamic sweep width interval data of the sub-region, and the maximum distance data of underground paths is obtained. The formula for calculating the distance between adjacent paths is as follows: in, This represents the maximum spacing data for underground paths; The overlap rate constraint interval; The angle from the beam center to the boundary; The terrain correction angle.

[0011] This invention achieves physical-driven optimization of the scanning path through the quantitative interaction of slope angle and acoustic parameters: First, based on the terrain sub-region dataset divided by contour lines (slope angle... Statistical calculations using TIN mesh normal vectors, excluding ±2σ outliers, were performed along with the multibeam opening angle. (Sound wave sector coverage angle) Constructing a dynamic sweep width formula This model employs a triple data coupling mechanism—the inverse cosine term. Quantization slope's geometric occlusion effect on beam boundaries (e.g.) This item makes (reduced by 38%) Correct for the projection loss of water depth along the slope direction. Associated device inherent performance – dynamic sweep width of output sub-region (in the Yangtze River case) (This improves accuracy by 72% compared to the traditional equidistant method); then, it is combined with a preset overlap rate constraint. From the spacing formula Generate the maximum allowable spacing where the numerator term Accurately characterize the minimum coverage requirement after terrain modulation (e.g.) When it increases Nonlinear decay drives steep slope regions (Compression), denominator term Then, the effective coverage potential of the device under slope constraints is defined, and the ratio of the two establishes a balance relationship between coverage intensity and coverage efficiency (when...). At that time, flat area (Up to 2.7 times the depth of the deep water area). This scheme was implemented in the Yangtze River waterway: inputting 38 sub-regions. Dataset (slope standard deviation) (Water depth range 3.5-53.5m), output spacing parameters after 0.8s calculation. The total length of the support path was optimized from 95.3km to 50.2km, and the void ratio was reduced from 9.2% to 0.3%. At the same time, the resolution non-uniformity index (deep water area resolution ratio) was compressed from 4.7 to 1.3, providing the optimal control framework at the data level for high-resolution topographic monitoring of water networks.

[0012] Preferably, step S3 includes the following steps: Step S31: Extract the centerline of the river channel based on the simulated underwater topography, and perform density zoning according to the preset slope angle threshold to obtain the waterway slope density zoning. Step S32: Generate terrain-adaptive paths for waterway slope density partitioning, and optimize path direction-endpoint connection to obtain terrain-adaptive paths; Step S33: Perform global overlap verification based on terrain adaptive path to obtain waterway adaptive scanning path.

[0013] This invention obtains the waterway centerline through contour line extraction or skeleton line algorithms, and performs density partitioning based on the slope angle calculation results of each path unit, dividing it into typical slope types such as gentle areas, transition areas, and steep slope areas, thereby establishing a spatial mapping relationship between slope gradient and path density. This density partitioning serves as a path control parameter, providing a spatially adaptive basis for spacing, radius of curvature, and sampling resolution in the subsequent path construction process. Subsequently, step S32 generates terrain-adaptive paths based on the partitioning results, using a dynamic direction adjustment algorithm to ensure maximum consistency between the path and the terrain's elevation trend in complex boundary areas, and combining path direction-endpoint connection optimization methods (such as Bezier endpoint correction and endpoint proximity constraints) to effectively avoid path breaks, intersections, and discontinuities. Step S33 introduces a global overlap rate verification method to calculate the coverage, overlap ratio, and redundant area ratio between different scanning paths, performing redundancy control and spatial layout balance optimization on the path set at a global scale, thereby forming a waterway adaptive scanning path set with high density control, good boundary fitting, and complete spatial coverage. This path set exhibits a spatial distribution characteristic that highly aligns with actual underwater topographic changes at the data representation level. This not only enhances the structural integrity of the scanning path but also provides strong data support for subsequent data acquisition efficiency, 3D modeling density, and reconstruction error control. Therefore, this invention, by constructing a terrain-adaptive scanning path generation mechanism based on slope density zoning, solves the problems of insufficient response to underwater slope structures, poor path continuity, and uncontrolled overlap rates in traditional path planning, thereby improving the spatial scanning efficiency and data acquisition accuracy of irregular underwater topographic areas.

[0014] Preferably, step S31 includes the following steps: Step S311: Calculate the river channel boundary TIN based on the irregular triangular network, and extract the river channel centerline from the simulated underwater topography. Step S312: Match the river channel boundary TIN with the river centerline to obtain the initial main path of the underwater river channel; Step S313: Based on the preset slope angle threshold, the initial main path of the underwater river channel is divided into density zones. The preset slope angle threshold is specifically set as follows: when the slope angle is less than 5°, it is divided into a flat zone; when the slope angle is greater than 5° and less than 15°, it is divided into a transition zone; when the slope angle is greater than 15°, it is divided into a steep slope zone; the flat zone and the transition zone are combined to generate the waterway slope density zone.

[0015] This invention generates a TIN structure based on pre-constructed underwater topographic simulation results. Through the node connections between triangular facets, it accurately depicts the microscopic topographic undulations and boundary contours of the river channel boundary. Based on this, a centerline extraction algorithm (such as skeleton extraction or terrain minimum resistance path analysis) is applied to extract the river channel centerline, forming the basic data support for the linear direction of the river channel. In step S312, the river channel boundary TIN and the centerline are matched using a spatial association method, that is, the minimum elevation path corresponding to the centerline is identified in the TIN, thereby generating an initial underwater main path with a realistic physical depression trend, avoiding the problem of traditional geometric centerlines ignoring actual hydrodynamic characteristics. Step S313 further extracts the local slope angle value of each segment of the initial main path based on slope angle calculation. Density is then divided according to a set threshold standard (<5° for flat areas, 5°~15° for transition areas, >15° for steep slope areas), forming a slope partition dataset with clear spatial differences. By combining flat and transitional zones into low-to-medium slope partitions, strategic adjustments can be made to path layout density and device scanning spacing. This allows subsequent path generation algorithms to adaptively adjust path control parameters based on terrain complexity, significantly reducing redundant path generation and coverage blind spots. This slope density partition dataset not only reflects the slope distribution characteristics of underwater terrain but also provides high-resolution data support for achieving spatial structure constraints and scanning efficiency optimization in path planning. Therefore, this invention, by constructing a TIN structure and introducing a slope angle partitioning algorithm, solves the problems of lagging perception of river slope changes, main path extraction deviating from the hydrodynamic center, and lack of terrain adaptability in path density settings in traditional methods, improving the accuracy of path structure representation and the level of data control refinement in complex river scenarios.

[0016] Preferably, step S32 includes the following steps: Step S321: Sampling control points for the waterway slope density zone at every 50m along the river centerline to obtain initial river control point data; Step S322: Fit the initial river control point data with uniform B-splines to obtain the initial river fitting data; Step S323: Based on the initial river channel fitting data, perform path direction optimization parallel to the river channel centerline correction to obtain the initial river channel topographic path; perform endpoint connection processing of adjacent areas using Bézier curves on the initial river channel topographic data to obtain the topographic adaptive path.

[0017] This invention extracts control points at 50-meter intervals along the river channel centerline to generate an initial river control point dataset. This operation not only ensures uniform sampling of the river channel morphology but also balances the efficiency control of terrain change capture and path generation at the data granularity level, providing basic node support for subsequent curve fitting. Subsequently, in step S322, the control point dataset is fitted using a uniform B-spline algorithm. Leveraging its advantages of local controllability and continuity, a smooth, self-intersecting fitted path is generated, effectively eliminating noise offsets caused by terrain disturbances or local abrupt changes in the original path, improving the geometric stability and analytical controllability of the centerline path in terrain representation. Following this, in step S323, path direction optimization is performed by using the fitted path as a central reference axis, ensuring that all scanned paths remain locally parallel to the fitted centerline direction, significantly improving the consistency of path organization, topological clarity, and the adaptability of the navigation algorithm. To further eliminate the problems of discontinuities and acute angle connections between path segments in different areas, the system uses Bézier curves to flexibly connect the path endpoints between zones. While maintaining topological constraints, it achieves high-order continuity connections, resulting in a globally smooth, locally controllable, and terrain-sensitive adaptive path structure. Through the combined processing of the above steps, path generation not only possesses high responsiveness to complex terrain but also controls the relationship between path density, orientation, and coherence in a data-driven manner, ensuring dynamic consistency between the path layout and the actual waterway topography. Therefore, this invention, through equidistant control point sampling, B-spline fitting, and Bézier curve connection, solves the problems of path breakage, curvature imbalance, and low terrain matching in traditional path generation processes, significantly improving the coherence, accuracy, and planning adaptability of underwater paths.

[0018] Preferably, step S33 includes the following steps: Global overlap verification is performed based on the terrain-adaptive path to obtain the waterway adaptive scan path. The correction formula for global overlap verification is as follows: in, The global average overlap rate; This represents the total number of terrain sub-regions. The water thickness of the i-th sub-region The angle of the acoustic wave fan-shaped coverage; This represents the total number of scan paths; For path range index; For the first The actual spacing between path intervals; For the first The average slope angle of the path interval.

[0019] This invention aggregates terrain sub-region parameters (slope angle) , water depth ) and measured path data (spacing) A quantitative balance model for coverage requirements and coverage capabilities was constructed—the numerator. The minimum cover requirement under complex terrain conditions affected by the slope-open angle coupling effect was accurately calculated (e.g., when...). When it increases The term attenuation highlights the need for enhanced coverage in steep slope areas, and the denominator term... Then, the effective coverage capability of the actual path layout is dynamically evaluated (wherein) Correcting the geometric cover loss caused by slope, (Related device performance); This model is achieved through The mathematical construction of this method, for the first time, achieves physically driven calculation of the global overlap rate (replacing empirical estimation), reducing the cavity prediction error from ±12% of the traditional method to ±2.3% in the Yangtze River case. Its data value is specifically manifested in three aspects: First, terrain adaptability—in the formula... and Spatial variability is explicitly coupled (e.g., in deep water areas) As the molecular value increases at time m, the driving path becomes more complex; steep slope region Firstly, by reducing the denominator value and automatically reducing the spacing, the coverage of the planned path in complex waterways (V / U / W-shaped cross-sections) is increased to 99.8%; secondly, computational efficiency is improved—only a single traversal is required. Sub-regions and The path range can be output (time complexity) In a 5km section of the Yangtze River The calculation time is less than 0.1 seconds, which is faster than Monte Carlo simulation. Thirdly, parameter sensitivity—through The deviation from the constraint interval [15%, 25%] (e.g.) ), directly generate path optimization instructions ( Time-distance compression ratio In the Yangtze River iteration, the initial... After two adjustments, the accuracy reached 18.3%, and the total path length was simultaneously optimized from 95.3km to 50.2km. Furthermore, the resolution unevenness (resolution ratio between deep and shallow water areas) was reduced from... .

[0020] Preferably, step S4 includes the following steps: Step S41: Perform a full-process path evaluation on the waterway adaptive scanning path and back-transmit the parameters of the adjacent path spacing calculation formula to obtain multi-source water network scanning path evaluation data. Step S42: Calculate the efficiency improvement ratio based on the waterway adaptive scanning optimization path and perform void ratio analysis to obtain full-process waterway analysis data; Step S43: Based on the full-process analysis data of the waterway and the multi-source water network scanning path evaluation data, plan the high-resolution, full-coverage scanning path of the water network multibeam bathymetry system.

[0021] This invention performs a full-process path evaluation on the adaptive scanning path of a waterway. The system collects and integrates multi-dimensional parameters such as spatial overlap rate, coverage density, turning change rate, and local path curvature during the path execution process to form a waterway planning path evaluation dataset. This dataset is fed back to the path generation module through a feedback mechanism, and parameters are adjusted in conjunction with indicators such as the average slope angle of the sub-region and scanning redundancy, thereby generating a more accurate and less redundant optimized adaptive scanning path for the waterway. In step S42, the efficiency improvement ratio of the optimized path execution is further analyzed. That is, by comparing the total path length, scanning time, number of control points, and other core indicators before and after optimization, the efficiency gain brought about by the path adjustment is quantified. At the same time, void rate analysis is performed in conjunction with terrain projection and path distribution data to identify uncovered areas during underwater scanning and locate their spatial distribution, constructing a spatial matrix of waterway void rate data, thereby providing spatial support for subsequent supplementary scanning strategies. In step S43, the path optimization results, efficiency change data, and void rate matrix are used as basic inputs. Multi-source topographic data, including water thickness distribution, underwater slope information, and riverbed geomorphological zoning, are comprehensively overlaid. The system automatically generates a high-resolution, full-coverage scanning path planning report and labels key planning sections, optimal path optimization areas, and suggested supplementary survey areas at the data level, forming a comprehensive topographic result file that can be used for engineering planning, scheduling management, and data visualization. Therefore, this process, by constructing a data-driven path evaluation and feedback mechanism, solves the problems of low coverage efficiency, weak data void identification capability, and difficulty in later result integration in traditional scanning path planning, significantly improving the closed-loop controllability of waterway scanning path planning and the comprehensive application value of the result data.

[0022] Preferably, step S41 includes the following steps: Step S411: Perform a full-process path evaluation on the waterway adaptive scanning path to obtain multi-source water network scanning path evaluation data; Step S412: Calculate the parameters of the average slope angle of the overlap rate constraint interval and the first sub-region from the waterway planning path evaluation data to obtain the waterway planning backhaul parameter set; Step S413: Adjust the calculation formula for the distance between adjacent paths based on the backhaul parameter set of the waterway planning to obtain the waterway adaptive scanning optimization path.

[0023] This invention performs a full-process path evaluation on the currently generated adaptive waterway scanning path. During the path execution simulation, the system collects key indicators such as path length, node curvature, redundancy overlap rate, and scan area boundary coverage density. It then combines this with river morphology information along the path to establish a "path-topography" corresponding dataset, thus forming waterway planning path evaluation data that can be used for quantitative analysis. Subsequently, step S412 refines the path evaluation data using sub-regions as basic units, focusing on extracting the constraint intervals of path overlap rate (such as maximum overlap rate, minimum coverage rate, etc.) and local terrain control parameters such as average slope angle within each sub-region, thereby forming a multi-dimensional planning feedback parameter set. This parameter set not only reflects the coupling relationship between path adaptability and terrain complexity but also provides a constraint basis for adaptive path optimization. In step S413, the system dynamically adjusts the control point sampling frequency, path smoothness, overlap tolerance, and scanning direction in the path generation logic in real time based on the feedback parameter set, making the optimized path more consistent with terrain change trends and maximizing coverage of the target area. During this process, all parameter adjustments are recorded in the path optimization change matrix to support multiple rounds of optimization and multi-time comparisons. In summary, this process constructs a closed-loop system of "data-driven—feedback optimization—path update" through structured extraction of path evaluation data, parameter feedback, and dynamic optimization control. This effectively solves the problems of insufficient response to terrain features and lack of accurate feedback in path adjustments in traditional waterway path design, improving the adaptability and execution efficiency of path planning. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the steps of a high-resolution topographic survey path planning method for water networks based on multi-source information. Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S3. Figure 3 for Figure 2 A schematic diagram of the adaptive scanning path of the waterway in step S33; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0025] The technical method of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0026] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0027] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0028] To achieve the above objectives, please refer to Figures 1 to 3 A high-resolution topographic survey path planning method for water networks based on multi-source information, the method comprising the following steps: Step S1: Collect multi-source historical data of the water network, integrate water level data from water level stations, and construct basic data on water body thickness; Step S2: Perform vertical mirroring on the basic dataset of water thickness and construct a vertical mirroring algorithm to output a water physical model; construct an irregular triangular network based on the water physical model and generate a simulated underwater terrain morphology; divide the simulated underwater terrain morphology into sub-regions and calculate the distance between adjacent paths to obtain the maximum distance data of underground paths. Step S3: Extract the river centerline based on the simulated underwater terrain morphology, and perform density partitioning according to the preset slope angle threshold to obtain waterway slope density partitioning; generate terrain adaptive paths for waterway slope density partitioning, and perform global overlap rate verification to obtain waterway adaptive scanning paths. Step S4: Perform a full-process path evaluation on the waterway adaptive scanning path to obtain multi-source water network scanning path evaluation data; based on the multi-source water network scanning path evaluation data, perform parameter feedback on the adjacent path spacing calculation formula to obtain the waterway planning feedback parameter set; based on the multi-source water network scanning path evaluation data and the waterway planning feedback parameter set, perform high-resolution, full-coverage scanning path planning for the water network multibeam echo sounding system.

[0029] In this embodiment of the invention, reference is made to Figure 1 The diagram shown is a flowchart illustrating the steps of a high-resolution topographic survey path planning method for water networks based on multi-source information according to the present invention. In this example, the high-resolution topographic survey path planning method for water networks based on multi-source information includes the following steps: Step S1: Collect multi-source historical data of the water network, integrate water level data from water level stations, and construct basic data on water body thickness; In this embodiment of the invention, multi-source basic data acquired from water level stations are fused and structured to construct a basic dataset of water body thickness with spatiotemporal consistency and the ability to express measurement depth features. During the data acquisition phase, the water level station, as a fixed observation facility, can continuously acquire static water level data for a specified section or region, and integrate hydrological parameters such as rainfall, flow velocity, and cross-sectional morphology. A continuous raw data stream is obtained by setting a sampling period. To achieve consistent fusion of multi-source information, the acquired data first needs to undergo spatiotemporal registration processing to ensure effective correspondence of observation data under a unified timestamp and geographic coordinate system. Subsequently, noise filtering and missing data interpolation are performed on the raw water level data, including threshold removal of abrupt outliers and smoothing using sliding window mean or empirical mode decomposition to improve data stability and resolvability. The calculation of water body thickness relies on the difference analysis between water level and local exposed terrain elevation data, combined with historical sounding data and a DEM (Digital Elevation Model) to perform point-by-point thickness calculation, generating water level profile thickness data, and constructing a thickness variation matrix across multiple time frames. Meanwhile, by integrating data from remote sensing imagery, GNSS measurement points, and watershed cross-section surveys, the spatial coverage and confidence level of thickness estimation are further improved. Ultimately, this process results in a multi-field structured basic dataset of water body thickness, including time labels, geographic coordinates, water level values, calculated thickness, and relevant meteorological boundary conditions, providing fundamental support for subsequent topographic inversion and triangulation modeling.

[0030] Step S2: Perform vertical mirroring on the basic dataset of water body thickness and construct a vertical mirroring algorithm to output a water body physical model; construct an irregular triangular network based on the water body physical model and generate a simulated underwater terrain morphology; divide the simulated underwater terrain morphology into sub-regions and calculate the distance between adjacent paths to obtain the maximum distance data of underground paths. In this embodiment of the invention, vertical mirror processing calculations, spatial structure grid construction, and terrain zoning feature extraction are performed based on a water thickness baseline dataset for topographic elevation inversion. First, the water thickness baseline dataset constructed in step S1 contains the following main fields: timestamp, water level measurement value, corresponding measurement point coordinates, estimated water thickness value, and local reference surface information. To achieve vertical mirror processing, the surface reference elevation for each measurement point must be determined, typically obtained through a DEM (Digital Elevation Model) or historical hydro-topographic maps. The vertical mirror processing formula uses the surface reference elevation minus the water thickness as its core calculation, and can be expressed as follows: ,in This represents the terrain elevation output by the vertical mirroring process. The original elevation datum value, This represents the water body thickness. After completing the vertical mirroring calculation of the elevation of all measuring points, the discretized terrain elevation points are constructed into a TIN (Triangular Irregular Network) according to their spatial distribution. Specifically, the Delaunay triangulation algorithm is used to connect the coordinate points, ensuring the continuity of the spatial structure and the accuracy of the contour feature representation. The generated TIN model can be directly used to construct the 3D simulation of underwater terrain and provide support for subsequent path extraction and scanning area division. After completing the 3D terrain model, the entire terrain surface is divided into sub-regions based on spatial layering and grid projection algorithms. Common methods include quadtree partitioning or contour spacing based on terrain curvature. The distance between adjacent paths is calculated within each sub-region, and the farthest distance between grid centerlines is obtained using the geodesic distance calculation method, forming the maximum distance data of underground paths. This lays the data foundation for density control and coverage analysis in subsequent path planning. The entire process requires that the data structure has a unified reference coordinate system in spatial distribution, and all processing results are output in the form of terrain nodes, triangular network topology, and regional spacing matrices.

[0031] Step S3: Extract the river centerline based on the simulated underwater terrain morphology, and perform density partitioning according to the preset slope angle threshold to obtain waterway slope density partitioning; generate terrain adaptive paths for waterway slope density partitioning, and perform global overlap rate verification to obtain waterway adaptive scanning paths. In this embodiment of the invention, based on the irregular triangular network (TIN) or its rasterized form constructed in step S2, gradient field analysis is performed on the terrain elevation surface. By calculating the local elevation difference and the rate of change of spatial coordinates, the slope information of the underwater terrain is extracted. Based on this, mainstream skeleton extraction algorithms (such as those based on MedialAxis or Voronoi diagrams) are used to extract the river centerline. This centerline represents the path of lowest potential energy in the terrain and can be considered as the natural runoff axis in the geomorphic evolution process. After obtaining the centerline, the entire terrain area is divided into multiple slope density segments according to a preset slope angle threshold. The specific calculation method includes performing sliding window statistics in the direction of the centerline normal to determine the frequency and numerical dispersion of slope changes per unit area, so as to achieve differentiated expression of high-variable areas and gentle slope areas. Subsequently, in the divided slope density zones, an adaptive path generation strategy is adopted for path layout according to different slope distribution characteristics. The core of this strategy is to introduce a spatial density adjustment function to automatically adjust the path spacing according to the slope fluctuation rate, ensuring that the path coverage matches the terrain complexity. Based on this, a global path overlap rate check is performed. Redundant, overlapping, or missing segments are identified using the Euclidean distance matrix between path segments and an angle cross-detection algorithm. The path set is then optimized overall, resulting in a scanned path with strong terrain responsiveness and coverage balance. Finally, all processing results are output in a unified format as structured data, including centerline coordinate sets, slope density grid data, adaptive path vector sets, and overlap rate analysis matrices, to support subsequent water network assessment and path optimization model calculations.

[0032] Step S4: Perform a full-process path evaluation on the waterway adaptive scanning path to obtain multi-source water network scanning path evaluation data; based on the multi-source water network scanning path evaluation data, perform parameter feedback on the adjacent path spacing calculation formula to obtain the waterway planning feedback parameter set; based on the multi-source water network scanning path evaluation data and the waterway planning feedback parameter set, perform high-resolution, full-coverage scanning path planning for the water network multibeam echo sounding system.

[0033] In this embodiment of the invention, a path attribute database is established based on the waterway adaptive scanning path vector set output in step S3. This database includes structured fields such as the start and end coordinates, length, direction angle, path spacing, turning angle variation, and coverage area number for each path. Subsequently, a path evaluation model is constructed to comprehensively analyze the path set in terms of overall coverage, local path redundancy, average overlap rate, scan connectivity, and backtracking path complexity. Coverage evaluation employs a path-region projection overlap algorithm to statistically analyze the overlap ratio between path trajectories and terrain sub-region grids. Redundancy analysis identifies spatial intersections and information duplication between path segments by calculating the minimum distance matrix and angular redundancy factor between paths. Simultaneously, the reachability and distribution uniformity of the global path system are verified using a path connectivity graph modeling approach, utilizing shortest path and connected component analysis from graph theory. Based on the path evaluation model, a parameter feedback mechanism is constructed. This mechanism normalizes the deviation range of various performance indicators and provides reverse adjustment suggestions for core parameters in the path generation strategy (such as path density function parameters, slope adjustment weights, and path turning limit angles), forming a feedback parameter set. This feedback mechanism guides parameter self-correction during the route replanning phase through iterative optimization. Finally, the route assessment data is integrated with terrain feature data to output a terrain structure representation of the multi-source water network, a route scan structure assessment table, and a spatial distribution layer by region. A full-coverage survey route planning report is constructed based on a unified data template, with output formats including vector data files (such as GeoJSON or Shapefile), tabular statistical results (CSV or Excel), and spatial graphical representations (PDF or WebGIS visualization data) to support subsequent scheduling simulations or construction planning analyses.

[0034] Preferably, step S1 includes the following steps: Step S11: Obtain the distance between upstream and downstream water level stations; deploy water level stations upstream and downstream of the target waterway, use pressure-type water level gauges to continuously observe the water level at a sampling frequency of 1Hz, and collect water level data from the water level stations. Step S12: Collect multi-source historical data of the water network, and collect single-beam bathymetry historical data from the multi-source historical data of the water network. Use the cross-section method to obtain the elevation of the underwater measuring point. Step S13: Grid the historical single-beam bathymetry data, generate interpolation data, and generate riverbed DEM data; Step S14: Synchronize the water level data from the water level stations with time, and use linear interpolation to calculate the water level data for the entire water area, taking into account the distance between upstream and downstream water level stations. Step S15: Spatially align the riverbed DEM data with the water surface elevation data of the entire water area to generate a basic dataset of water thickness.

[0035] In this embodiment of the invention, pressure level gauges are deployed upstream and downstream of the target waterway, and continuous high-frequency water level observations are performed at a sampling frequency of 1Hz to acquire time-series high-resolution water surface height data from the level stations. A GPS timing module is configured to ensure accurate and consistent time stamps. Simultaneously, the distance between level stations is collected as a spatial constraint for linear interpolation, constraining the boundary conditions for water surface elevation calculation. Subsequently, multi-source historical data of the water network are collected, along with the most recent single-beam bathymetry historical data for the river section. Combined with cross-sectional information from the measurement route, the cross-sectional method is used to extract the coordinates of the measuring points and the corresponding water depth from the bathymetry echoes, and the riverbed elevation point cloud data is derived by overlaying the water surface elevation at that time. When spatially gridding this dataset, a regular grid or TIN structure is introduced, and raster reconstruction is performed at a resolution of 1m or higher. Inverse distance weighted interpolation (IDW) or spline interpolation algorithms (such as Spline) are used to interpolate the point cloud data, generating a continuous riverbed DEM (Digital Elevation Model). This model represents the topographic elevation value of each grid point. To obtain full-area water surface elevation data, the water level observation data from the previous steps needs to be time-synchronized. Water level values ​​at each observation point are identified under a unified time reference. Based on the distance between upstream and downstream water level stations and the instantaneous water level data at both ends, linear interpolation is used to spatially reconstruct the water surface elevation distribution at various locations along the entire river section. This calculation process employs an equally spaced segmentation strategy, mapping the water level elevation to coordinates under the DEM grid projection. Finally, the water surface elevation grid data is spatially aligned with the riverbed DEM, i.e., point-by-point matching is performed under the same projection system and coordinate reference. The difference between the water surface elevation and the riverbed elevation at each grid point is calculated, yielding thickness data reflecting the water depth in different areas. This multivariate dataset, comprising X and Y coordinates, water surface elevation, riverbed elevation, and water thickness, possesses spatial continuity and temporal synchronization, serving as the input basis for subsequent vertical topographic mirroring and scanning path planning.

[0036] Of particular importance is the need to pay attention to step S13: The minimum triangle angle for gridding should be greater than 25°. When detecting bank slopes and deep pits, the density of sampling should be modified to a spacing of 0.5m.

[0037] In this embodiment of the invention, the accuracy of terrain modeling is improved through a constraint algorithm: First, when constructing the Triangular Irregular Network (TIN), a Delaunay subdivision algorithm with angle constraints is used, which forces that the minimum interior angle of all triangles be >25° (traditional algorithms usually only >10°). Specifically, this is implemented by iteratively detecting triangles with excessively small angles—if an interior angle <25° is detected, a Steiner point is inserted at the centroid of the circumcircle of the triangle and the triangle is retriangulated until all elements satisfy the angle constraints (in the Yangtze River case, an average of 7.2 iterations were performed, adding 12.3% of auxiliary points); Second, for areas with abrupt changes in terrain features (bank slopes are defined as areas with a slope change rate >5° / 10m, and deep pits are defined as...),... For closed depressions with a depth > 5m and a curvature of 0.1m⁻¹, an adaptive densification sampling mechanism is initiated: After identifying the feature area boundary based on the initial TIN grid, densification grid points are generated at 0.5m intervals along the boundary normal direction (densification along the slope direction for bank slopes, and concentric circles radially for deep pits). These densification points are then forcibly added as fixed nodes to the triangulation network reconstruction. For example, in a steep bank slope area of ​​the Yangtze River (original point spacing 2m), 327 densification points are inserted to improve the local resolution to 0.5m. In deep pit areas, after extracting the boundary using 0.5m isobaths, densification points are placed every 22.5° azimuth along the pit wall. This process requires coupling a terrain curvature detection algorithm—using the moving least squares method to fit the local surface and calculate the Gaussian curvature. ,when Encryption is triggered on time, and normal vector mutation analysis (angle between normal vectors of adjacent triangular faces > 30°) is used to assist in identifying bank slope inflection lines. The final output is a TIN model that meets both geometric quality and topographic fidelity standards, providing a basic data structure for millimeter-level error control in subsequent water depth inversion (Yangtze River validation shows that the elevation interpolation error in the bank slope area is...). (Decreased to 0.11m).

[0038] Preferably, in step S2, the basic dataset of water body thickness is mirrored vertically, and the formula for vertical mirroring is constructed as follows: The basic dataset of water body thickness is mirrored in the vertical direction to obtain mirrored data of water body thickness in the vertical direction. A vertical mirror processing algorithm is constructed based on the mirror processing data of the water body thickness in the vertical direction to obtain a physical model of the water body. The formula for vertical mirror processing is as follows: in, Let be the planar coordinates of any point in the waterway; For point Riverbed elevation data at the location; For point Water surface elevation data at the location; This represents the real-time water level at upstream water level station A. This is the output physical model of the water body.

[0039] In this embodiment of the invention, the basic dataset of water body thickness constructed in step S1 includes the following main fields: timestamp, water level measurement value, corresponding measuring point coordinates, estimated water body thickness value, and local reference surface information. To achieve vertical mirroring, the surface reference elevation of each measuring point must first be determined, typically obtained through a DEM (Digital Elevation Model) or historical hydrological topographic map. The vertical mirroring formula uses the surface reference elevation minus the water body thickness as its core calculation, and can be expressed as follows: ,in This represents the terrain elevation output by the vertical mirroring process. The original elevation datum value, This represents the water body thickness. After completing the vertical mirroring calculation of the elevation of all measuring points, the discretized terrain elevation points are constructed into a TIN (Triangular Irregular Network) according to their spatial distribution. Specifically, the Delaunay triangulation algorithm is used to connect the coordinate points, ensuring the continuity of the spatial structure and the accuracy of the contour feature representation. The generated TIN model can be directly used to construct the 3D simulation of underwater terrain and provide support for subsequent path extraction and scanning area division. After completing the 3D terrain model, the entire terrain surface is divided into sub-regions based on spatial layering and grid projection algorithms. Common methods include quadtree partitioning or contour spacing based on terrain curvature. The distance between adjacent paths is calculated within each sub-region, and the farthest distance between grid centerlines is obtained using the geodesic distance calculation method, forming the maximum distance data of underground paths. This lays the data foundation for density control and coverage analysis in subsequent path planning. The entire process requires that the data structure has a unified reference coordinate system in spatial distribution, and all processing results are output in the form of terrain nodes, triangular network topology, and regional spacing matrices.

[0040] Preferably, step S2, which involves dividing the simulated underwater terrain into sub-regions and calculating the distance between adjacent paths, includes: The simulated underwater topography is divided into sub-regions according to the contour lines, and the average slope angle of the sub-regions is calculated to obtain the underwater multibeam bathymetry opening angle data. Dynamic sweep width is calculated based on underwater multibeam bathymetry opening angle data to obtain dynamic sweep width interval data for sub-regions. The formula for calculating dynamic sweep width is as follows: Among them, For the i-th sub-region, the sub-region dynamic sweep width interval data; Let be the average slope angle of the i-th sub-region; The angle of the acoustic wave fan-shaped coverage; Let be the water thickness of the i-th sub-region; The distance between adjacent paths is calculated based on the dynamic sweep width interval data of the sub-region, and the maximum distance data of underground paths is obtained. The formula for calculating the distance between adjacent paths is as follows: in, This represents the maximum spacing data for underground paths; The overlap rate constraint interval; The angle from the beam center to the boundary; The terrain correction angle.

[0041] In this embodiment of the invention, terrain feature quantization is coupled with an acoustic coverage model: firstly, the terrain model is processed based on vertical mirroring. The bottom boundary TIN is used to extract contour lines (contour interval is usually set to 0.5-2m), dividing the waterway into sections. For each of the homogeneous sub-regions, calculate the average slope angle. —This parameter is obtained by statistically analyzing the median of the tilt angles of the normal vectors of each grid face in the Delaunay triangulation (excluding ±2σ outliers), forming a slope angle distribution dataset; this is then combined with the inherent opening angle parameter of the multibeam system. (sonic wave sector coverage angle) and average water depth in sub-regions (Derived from a 3D model of water body thickness), applying a dynamic sweep width formula Calculate the effective coverage width, where the inverse cosine term is used. The geometric interference effect of terrain slope and beam boundary was quantified, and The basic coverage capability of water depth and equipment parameters was characterized; subsequently, based on the preset overlap rate constraint interval... Through the formula for the distance between adjacent paths Calculate the maximum allowable spacing, here the numerator term The minimum cover requirement after slope correction is the denominator term. This indicates the effective coverage potential of the equipment under the influence of slope; the ratio of these two values ​​precisely balances the constraints between terrain complexity and survey efficiency. The entire process was automated in the Yangtze River case study: 897,542 TIN grid points were input to generate 38 sub-regions. The calculation took 0.8 seconds (accuracy ±0.3°), and the output was a dynamic sweep width. and maximum spacing This provides a quantitative basis for subsequent path optimization.

[0042] As an example of the present invention, reference is made to Figure 2 As shown, step S3 in this example includes: Step S31: Extract the centerline of the river channel based on the simulated underwater topography, and perform density zoning according to the preset slope angle threshold to obtain the waterway slope density zoning. Step S32: Generate terrain-adaptive paths for waterway slope density partitioning, and optimize path direction-endpoint connection to obtain terrain-adaptive paths; Step S33: Perform global overlap verification based on terrain adaptive path to obtain waterway adaptive scanning path.

[0043] In this embodiment of the invention, based on the previously constructed simulated underwater topography, i.e., an irregular triangular network (TIN) or raster topographic surface model with topographic elevation values, the key to extracting the river centerline lies in identifying the path of lowest potential energy in the topography. Specifically, the lowest depression region of the riverbed is first obtained through elevation inversion. Then, the centerline point set is extracted based on the centroid tracking method or watershed partitioning algorithm using isoelastic elevation zones, forming the centerline vector path of the dominant water flow direction. This centerline serves as the main axis of spatial analysis. Isochronous sampling is performed on both sides of its normal direction. The local slope angle is calculated using the elevation change rate near the centerline. Based on a preset slope angle threshold, the entire centerline is segmented to generate a slope density partition dataset, which includes the spatial range, average slope value, change rate statistics, and segment identification code for each path segment. Next, terrain-adaptive path generation is performed within each slope density zone. The specific strategy involves setting a function adjustment coefficient for path spacing based on slope intensity, setting high-density path coverage in high-slope areas and sparse path layout in low-slope areas. Then, a path direction-endpoint optimization algorithm is introduced to analyze the directional consistency of the initially generated path segments. Directional smoothing is achieved through vector angle judgment and a dynamic window moving average method. Simultaneously, minimum distance matching between path endpoints and adjacent paths is performed to achieve global path coherence and continuous segment connection. Finally, spatial overlap analysis is performed on the entire generated path set. A path segment adjacency matrix is ​​constructed using the path vector overlay method. An overlap discrimination function is built using the Euclidean distance between paths and the angle between segments to identify highly redundant or missing areas between path segments. Path elimination or completion optimization is performed based on the overlap rate threshold, ultimately generating a spatially complete and structurally continuous waterway adaptive scan path dataset. Its output format includes centerline vector data, slope zone attribute table, adaptive path segment set, and overlap rate matrix.

[0044] Preferably, step S31 includes the following steps: Step S311: Calculate the river channel boundary TIN based on the irregular triangular network, and extract the river channel centerline from the simulated underwater topography. Step S312: Match the river channel boundary TIN with the river centerline to obtain the initial main path of the underwater river channel; Step S313: Based on the preset slope angle threshold, the initial main path of the underwater river channel is divided into density zones. The preset slope angle threshold is specifically set as follows: when the slope angle is less than 5°, it is divided into a flat zone; when the slope angle is greater than 5° and less than 15°, it is divided into a transition zone; when the slope angle is greater than 15°, it is divided into a steep slope zone; the flat zone and the transition zone are combined to generate the waterway slope density zone.

[0045] In this embodiment of the invention, a three-dimensional elevation model simulating underwater topography is used to construct a TIN model of the river channel region using the Delaunay triangulation algorithm, and its boundary control conditions are defined to accurately reflect the irregular geometric features of the underwater topographic boundary. By calculating the local normal vector and identifying the boundary gradient direction of each triangular unit in the TIN model, the boundary curve of the river channel is determined and a river channel boundary TIN is formed. This boundary TIN represents the closed edge structure of the river channel in three-dimensional space. Subsequently, based on existing centerline extraction methods (such as contour centroid tracing or the shortest flow path method), low-potential paths are identified in the TIN elevation network, and continuous centerline vector data is extracted. To further improve the fit between the path and the actual river channel, a deep channel line matching algorithm is introduced. The centerline coordinates and the topographic channel line formed by the boundary TIN are matched by minimum distance and the elevation gradient direction is analyzed for consistency. By calculating the vertical distance difference between the centerline point and the TIN node and the slope vector angle, the path segment that matches the deep channel line is identified, and finally the initial main path of the underwater river channel is determined. Based on this, continuous slope angle calculations are performed on the main path. A sliding window differential algorithm is used to calculate the slope value based on the elevation difference and horizontal distance between each pair of path nodes. The resulting slope data is then mapped to the main path vector attribute table to form a slope sequence. The main path is divided into intervals according to preset slope angle thresholds: slope angles < 5° are defined as flat areas, slopes between 5° and 15° are defined as transition areas, and slopes greater than 15° are defined as steep areas. The division logic can be implemented using a multi-threshold piecewise function, and path nodes are attribute-encoded using partition labels. To optimize subsequent path generation strategies, flat and transition areas are merged and uniformly set as gently sloping areas, thus constructing a spatially continuous waterway slope density partitioning layer with clear hydraulic classification. The output includes main path vector data, node slope sequences, density partitioning encoding tables, and spatial partitioning vector layers, which serve as the basis for subsequent adaptive path design.

[0046] Preferably, step S32 includes the following steps: Step S321: Sampling control points for the waterway slope density zone at every 50m along the river centerline to obtain initial river control point data; Step S322: Fit the initial river control point data with uniform B-splines to obtain the initial river fitting data; Step S323: Based on the initial river channel fitting data, perform path direction optimization parallel to the river channel centerline correction to obtain the initial river channel topographic path; perform endpoint connection processing of adjacent areas using Bézier curves on the initial river channel topographic data to obtain the topographic adaptive path.

[0047] In this embodiment of the invention, based on the constructed waterway slope density zoning layer, equidistant sampling is first performed along the centerline direction at fixed intervals of 50m. The spatial coordinates of the corresponding positions on the centerline are extracted as control points, forming an initial river control point dataset. This control point set expresses the path spatial structure in two-dimensional or three-dimensional coordinates (X,Y,H), and its slope zoning category is labeled for subsequent structure discrimination. After the control point sampling is completed, a uniform B-spline curve fitting algorithm is used to smooth the control point sequence. The B-spline fitting process introduces node vectors and a set of basis functions, and generates each curve segment through recursive construction, ensuring that the curve is geometrically continuous and satisfies the C² smoothness condition. The fitting output is the initial river fitting data, which not only maintains the overall trend of the original path but also effectively filters out path abrupt changes caused by local topographic fluctuations. Next, path direction optimization is performed based on the fitted curve. Specifically, the tangent vector direction of the curve is calculated on each spline segment, and it is used as the reference direction to perform parallel offset correction on subsequent path segments, thereby ensuring that the generated path direction is consistent with the overall direction of the centerline and avoiding path direction deviation due to fitting errors. When there are discontinuous regions with spatial breaks or partition boundaries between path segments, Bézier curves are further introduced for endpoint connection processing. Cubic Bézier curves are used to control the endpoint coordinates and tangent vector direction, completing the continuous connection of adjacent path segments. This connection process must ensure the continuity of curvature changes and uniformly distribute the node parameters t in the parameter space, thereby forming a composite path with a certain degree of physical rationality and geometric smoothness in the connection transition section. The final terrain-adaptive path includes a spatial coordinate sequence, path direction vector field, fitting error statistics, and curvature distribution data, which can be output as vector line data and attribute table structure to support subsequent global overlap rate analysis and spatial optimization operations of the scanned path.

[0048] Preferably, step S33 includes the following steps: Global overlap verification is performed based on the terrain-adaptive path to obtain the waterway adaptive scan path. The correction formula for global overlap verification is as follows: in, The global average overlap rate; This represents the total number of terrain sub-regions. The water thickness of the i-th sub-region The angle of the acoustic wave fan-shaped coverage; This represents the total number of scan paths; For path range index; For the first The actual spacing between path intervals; For the first The average slope angle of the path interval.

[0049] In this embodiment of the invention, the global overlap verification technique is achieved through spatial aggregation calculation of multi-source terrain parameters and measured path data: First, based on the dataset of m divided terrain sub-regions (containing the average water depth of each region) With slope angle ) and the measured spacing data of the planned n scanning paths Perform spatial topology matching—match each path interval Mapping to its covered sub-regions, a path-terrain correlation matrix is ​​established; then, a correction formula is applied. Perform global overlap rate calculation, where the numerator term It is the sum of the minimum coverage requirements for all sub-regions (water depth) Determine the base cover intensity; double cosine terms quantify terrain slope. Angle with equipment Interference attenuation effect), denominator term This represents the total effective coverage capability (path spacing) of the actual path layout. Slope Angle The correction reflects the actual coverage width. (Related equipment performance); data coupling relationships must be strictly handled during calculation: slope angle Take the sub-region covered by the path interval The weighted average (weighted by the ratio of the projected area of ​​the sub-region within the interval), open angle Attenuation compensation needs to be performed based on water depth. (To correct for sound wave propagation loss); after completing the initial calculation, verify... Does it fall within the preset range [15%, 25%]? If it exceeds this range, iterative optimization is triggered. For path intervals where the spacing exceeds the limit Perform binary compression (step size 5), when Time-merging spacing redundancy path (And the radius of curvature > 200m), after each iteration, the path-terrain topology is re-associated and the formula parameters are updated until the constraints are met. In the Yangtze River network verification, the initial... After two iterations (compressing the spacing between 3 paths and merging 1 redundant path), the accuracy was improved to 18.3%, and the final adaptive waterway scanning path was output.

[0050] Of particular importance is that when performing global overlap verification, the following should also be noted: When the water depth is greater than 40m, the spacing between paths should be increased to 1.2m. ; When the water depth is less than 5m, the path spacing should be reduced to 0.8m. ; The radius of curvature at the inflection point in the calculation should be greater than or equal to 3. .

[0051] In this embodiment of the invention, refined operation control is achieved through water depth grading and path geometric constraints: First, in the path planning stage, three-dimensional model data of water thickness are integrated in real time, and the water depth of each path interval is controlled accordingly. Perform threshold detection—when At time m (deep water area), the theoretical maximum spacing will be Multiply by the expansion factor of 1.2 (i.e.) This operation is based on the energy attenuation formula of sound waves in deep water. ( (for frequency), increasing the spacing can compensate for the increase in effective sweep width caused by energy attenuation; when When the depth is m (shallow water area), a compressibility factor of 0.8 should be applied. This is to compensate for coverage loss caused by edge beam angle distortion in multi-beam systems in shallow water (experience shows that the actual coverage width is reduced by about 20% in shallow water). Secondly, in the path geometry generation stage, the radius of curvature for each path is... Implement hard constraints: control point coordinates via B-spline path Calculate local curvature When detected (Right now At the turning point of curvature, the Clothoid spiral smoothing algorithm is activated—extending before and after the curvature abrupt change point. Length of the transition segment ( (For the minimum turning radius of the equipment), and resolve the control point coordinates until the condition is met. (The maximum curvature in the Yangtze River case is 0.15m) (Decreased to 0.03m⁻¹). This process requires coupling with a real-time depth data stream: updating the pathpoint depth values ​​at a frequency of 10Hz. (Derived by bilinear interpolation of the gridded water thickness model), the spacing coefficient is dynamically adjusted; at the same time, the path curvature is solved by the Frenet frame coordinate system, and additional smoothing points are inserted in the deep-water-shallow water transition zone (such as the edge of a deep pit where H suddenly changes from 45m to 3m) to avoid sudden changes in equipment attitude, and finally an optimized path that conforms to the laws of underwater acoustic physics and kinematic constraints is formed.

[0052] Preferably, step S4 includes the following steps: Step S41: Perform a full-process path evaluation on the waterway adaptive scanning path and back-transmit the parameters of the adjacent path spacing calculation formula to obtain multi-source water network scanning path evaluation data. Step S42: Calculate the efficiency improvement ratio based on the waterway adaptive scanning optimization path and perform void ratio analysis to obtain full-process waterway analysis data; Step S43: Based on the full-process analysis data of the waterway and the multi-source water network scanning path evaluation data, plan the high-resolution, full-coverage scanning path of the water network multibeam bathymetry system.

[0053] In this embodiment of the invention, an optimal path set is iteratively generated through full-process evaluation feedback of the adaptive scanning path, and a terrain planning output for multi-source water networks is formed by combining spatial distribution feature analysis. First, for the generated adaptive scanning paths of waterways, a path performance evaluation index system is constructed. This system includes core parameters such as the area covered per unit scan length, path overlap rate, average path curvature rate, and path density distribution dispersion. This parameter system is used to perform a full-process analysis of the scanning path. During the evaluation process, the path vector line set is used as input, combined with a high-resolution underwater terrain grid model, and path buffer analysis technology is used to count the number of effective grid cells that each path can cover, and construct an efficiency ratio parameter by comparing it with the path length. Simultaneously, the redundancy and missed scan risk between paths are evaluated using the minimum distance matrix and overlap interval mapping between adjacent paths, and the average overlap rate between paths is spatially weighted to identify inefficient or high-density areas in the path planning. Subsequently, the evaluation results were processed by parameter feedback. Path segments identified as inefficient or omitted in the performance evaluation were used as feedback input. A local path reconstruction operation was performed using the original terrain model. The path was updated through a shortest feasible distance algorithm and a region densification generation mechanism, ultimately forming a set of adaptive scanning optimized paths for the waterway. Based on the optimized paths, an efficiency improvement ratio was calculated. By comparing the path coverage efficiency parameters before and after optimization, the performance change after path planning optimization was quantitatively calculated. Furthermore, based on the raster mapping results of the optimized paths in space, a void ratio analysis was performed. The void ratio is defined as the proportion of grid cells not covered by any scanning path to the total terrain area grid cells. This was achieved using spatial overlay analysis combined with binary mask calculation. The analysis results were output as a void heat map and void ratio statistics. Based on the full-process waterway analysis data, a structured terrain planning report was generated. The report included a path parameter evaluation table, an efficiency ratio evolution diagram, a void ratio distribution map, and an optimized path spatial layout map. A template-based filling mechanism was used to output a visually appealing and readable full-coverage scanning path planning report, providing data support for subsequent underwater operation deployment and path control strategies.

[0054] Preferably, step S41 includes the following steps: Step S411: Perform a full-process path evaluation on the waterway adaptive scanning path to obtain multi-source water network scanning path evaluation data; Step S412: Calculate the parameters of the average slope angle of the overlap rate constraint interval and the first sub-region from the waterway planning path evaluation data to obtain the waterway planning backhaul parameter set; Step S413: Adjust the calculation formula for the distance between adjacent paths based on the backhaul parameter set of the waterway planning to obtain the waterway adaptive scanning optimization path.

[0055] In this embodiment of the invention, the adaptive scanning path of the waterway is optimized and adjusted by constructing an evaluation dataset and a feedback parameter set. In step S411, based on the existing vector dataset of the adaptive scanning path of the waterway, each scanning path is first buffered with a certain width using a path buffer analysis method to form a path influence zone; then, it is spatially overlaid with an underwater topographic raster model, and the number of times each grid cell is covered by the scanning path is counted to calculate the path overlap rate matrix, identifying high redundancy areas with an overlap number greater than a threshold and void areas with an overlap number less than 1; in addition, by analyzing the angle between the path direction vector field and the topographic slope direction, the consistency between the path direction and the topographic trend can be extracted to further construct a path deviation index. The spatial coverage characteristics, overlap distribution, and direction deviation of all path segments are encoded into the path attribute table to form the waterway planning path evaluation data. In step S412, the waterway planning path evaluation data is partitioned and aggregated by sub-regions. By presetting sub-region boundaries or performing spatial clustering based on a slope density layer, the average overlap rate and average slope angle parameters of the path segments within each sub-region are calculated. Overlap rate statistics are calculated using the ratio of the area repeatedly covered within a region to the total scanned area. The average slope angle is obtained based on the slope map calculated from the DEM raster using a regional weighted average method, and compared with the average elevation change of the path segments to form backhaul parameters that can be used to adjust the scan path density and orientation. Finally, in step S413, parameter backhaul adjustments are performed based on the waterway planning backhaul parameter set. Specifically, a dynamic adjustment function is introduced through the path reconstruction module to locally compress or expand the path segment spacing, and the path orientation is adjusted based on the average slope angle information to better conform to the natural slope trend of the underwater terrain. This process is implemented by combining the Bézier curve reconstruction mechanism and the path node direction vector correction algorithm. The output waterway adaptive scan optimization path has better overlap distribution, slope adaptability, and scan efficiency. The path dataset includes an updated spatial attribute table, backhaul adjustment markers, and regional performance statistics, providing a foundation for subsequent path efficiency analysis and terrain planning.

[0056] The beneficial effects of this invention are that it enables continuous monitoring of upstream and downstream water levels through the deployment of water level stations. After collecting multi-source water level data, it synchronizes historical depth data and generates a basic dataset of water body thickness. This dataset has both temporal consistency and spatial correspondence, providing accurate boundary conditions for subsequent inversion modeling. In step S2, the water body thickness data is mirrored in the vertical direction of the terrain, and an irregular triangular network (TIN) is constructed in combination with elevation differences. This accurately reproduces the spatial undulations of the underwater landform, especially showing good terrain restoration capabilities in irregular river areas. Based on this, by dividing the TIN model into sub-regions and calculating path spacing, the maximum spacing data between the scanning blind zone and path coverage is effectively identified, thus providing a quantitative basis for the construction of subsequent path distribution strategies. Step S3 further divides the river centerline extracted from the triangular network model into slope angle partitions, dividing areas with different degrees of terrain change. Based on density characteristics, terrain-adaptive paths are automatically generated, and a global overlap rate verification mechanism is introduced to optimize scanning uniformity and coverage integrity from the perspective of path spatial layout. Step S4 introduces a path evaluation model to diagnose the performance of all scanned paths and feeds back evaluation parameters such as overlap rate, void rate, and path efficiency to guide the iterative path adjustment process and achieve adaptive optimization of path planning. Furthermore, a terrain planning report is automatically generated based on the multi-source water network scanned path evaluation data, visually presenting the path-terrain-coverage relationship and providing a quantitative decision-making basis for underwater operations.

[0057] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for high-resolution topographic surveying path planning of water networks based on multi-source information, characterized in that, Includes the following steps: Step S1: Collect multi-source historical data of the water network, integrate water level data from water level stations, and construct basic data on water body thickness; Step S2: Perform vertical mirroring on the basic dataset of water body thickness and construct a vertical mirroring algorithm to output a water body physical model; construct an irregular triangular network based on the water body physical model and generate a simulated underwater terrain morphology; divide the simulated underwater terrain morphology into sub-regions and calculate the distance between adjacent paths to obtain the maximum distance data of underground paths. Step S3: Extract the river centerline based on the simulated underwater terrain morphology, and perform density partitioning according to the preset slope angle threshold to obtain waterway slope density partitioning; generate terrain adaptive paths for waterway slope density partitioning, and perform global overlap rate verification to obtain waterway adaptive scanning paths. Step S4: Perform a full-process path evaluation on the waterway adaptive scanning path to obtain multi-source water network scanning path evaluation data; Based on the multi-source water network scanning path evaluation data, the parameters for calculating the distance between adjacent paths are fed back to obtain the waterway planning backhaul parameter set; based on the multi-source water network scanning path evaluation data and the waterway planning backhaul parameter set, high-resolution, full-coverage scanning path planning for the water network multibeam bathymetry system is carried out.

2. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain the distance between upstream and downstream water level stations; deploy water level stations upstream and downstream of the target waterway, use pressure-type water level gauges to continuously observe the water level at a sampling frequency of 1Hz, and collect water level data from the water level stations. Step S12: Collect multi-source historical data of the water network, and collect single-beam bathymetry historical data from the multi-source historical data of the water network. Use the cross-section method to obtain the elevation of the underwater measuring point. Step S13: Grid the historical single-beam bathymetry data, generate interpolation data, and generate riverbed DEM data; Step S14: Synchronize the water level data from the water level stations with time, and use linear interpolation to calculate the water level data for the entire water area, taking into account the distance between upstream and downstream water level stations. Step S15: Spatially align the riverbed DEM data with the water surface elevation data of the entire water area to generate a basic dataset of water thickness.

3. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 1, characterized in that, Step S2 involves mirroring the basic water thickness dataset vertically and constructing the vertical mirroring formula, including: The basic dataset of water body thickness is mirrored in the vertical direction to obtain mirrored data of water body thickness in the vertical direction. A vertical mirror processing algorithm is constructed based on the mirror processing data of the water body thickness in the vertical direction to obtain a physical model of the water body. The formula for vertical mirror processing is as follows: in, Let be the planar coordinates of any point in the waterway; For point Riverbed elevation data at the location; For point Water surface elevation data at the location; This represents the real-time water level at upstream water level station A. This is the output physical model of the water body.

4. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 1, characterized in that, Step S2 involves dividing the simulated underwater terrain into sub-regions and calculating the distance between adjacent paths, including: The simulated underwater topography is divided into sub-regions according to the contour lines, and the average slope angle of the sub-regions is calculated to obtain the underwater multibeam bathymetry opening angle data. Dynamic sweep width is calculated based on underwater multibeam bathymetry opening angle data to obtain dynamic sweep width interval data for sub-regions. The formula for calculating dynamic sweep width is as follows: Among them, For the i-th sub-region, the sub-region dynamic sweep width interval data; Let be the average slope angle of the i-th sub-region; The angle of the acoustic wave fan-shaped coverage; Let be the water thickness of the i-th sub-region; The distance between adjacent paths is calculated based on the dynamic sweep width interval data of the sub-region, and the maximum distance data of underground paths is obtained. The formula for calculating the distance between adjacent paths is as follows: in, This represents the maximum spacing data for underground paths; The overlap rate constraint interval; The angle from the beam center to the boundary; The terrain correction angle.

5. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Extract the centerline of the river channel based on the simulated underwater topography, and perform density zoning according to the preset slope angle threshold to obtain the waterway slope density zoning. Step S32: Generate terrain-adaptive paths for waterway slope density partitioning, and optimize path direction-endpoint connection to obtain terrain-adaptive paths; Step S33: Perform global overlap verification based on terrain adaptive path to obtain waterway adaptive scanning path.

6. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 5, characterized in that, Step S31 Includes the following steps: Step S311: Calculate the river channel boundary TIN based on the irregular triangular network, and extract the river channel centerline from the simulated underwater topography. Step S312: Match the river channel boundary TIN with the river centerline to obtain the initial main path of the underwater river channel; Step S313: Based on the preset slope angle threshold, the initial main path of the underwater river channel is divided into density zones. The preset slope angle threshold is specifically set as follows: when the slope angle is less than 5°, it is divided into a flat zone; when the slope angle is greater than 5° and less than 15°, it is divided into a transition zone; when the slope angle is greater than 15°, it is divided into a steep slope zone; the flat zone and the transition zone are combined to generate the waterway slope density zone.

7. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 5, characterized in that, Step S32 includes the following steps: Step S321: Sampling control points for the waterway slope density zone at every 50m along the river centerline to obtain initial river control point data; Step S322: Fit the initial river control point data with uniform B-splines to obtain the initial river fitting data; Step S323: Based on the initial river channel fitting data, perform path direction optimization parallel to the river channel centerline correction to obtain the initial river channel topographic path; perform endpoint connection processing of adjacent areas using Bézier curves on the initial river channel topographic data to obtain the topographic adaptive path.

8. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 5, characterized in that, Step S33 includes the following steps: Global overlap verification is performed based on the terrain-adaptive path to obtain the waterway adaptive scan path. The correction formula for global overlap verification is as follows: in, The global average overlap rate; This represents the total number of terrain sub-regions. Let be the water thickness of the i-th sub-region; The angle of the acoustic wave fan-shaped coverage; This represents the total number of scan paths; For path range index; For the first The actual spacing between path intervals; For the first The average slope angle of the path interval.

9. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Perform a full-process path evaluation on the waterway adaptive scanning path and back-transmit the parameters of the adjacent path spacing calculation formula to obtain multi-source water network scanning path evaluation data. Step S42: Calculate the efficiency improvement ratio based on the waterway adaptive scanning optimization path and perform void ratio analysis to obtain full-process waterway analysis data; Step S43: Based on the full-process analysis data of the waterway and the multi-source water network scanning path evaluation data, plan the high-resolution, full-coverage scanning path of the water network multibeam bathymetry system.

10. The method for high-resolution topographic surveying path planning of water networks based on multi-source information according to claim 9, characterized in that, Step S41 includes the following steps: Step S411: Perform a full-process path evaluation on the waterway adaptive scanning path to obtain multi-source water network scanning path evaluation data; Step S412: Calculate the parameters of the average slope angle of the overlap rate constraint interval and the first sub-region from the waterway planning path evaluation data to obtain the waterway planning backhaul parameter set; Step S413: Adjust the calculation formula for the distance between adjacent paths based on the backhaul parameter set of the waterway planning to obtain the waterway adaptive scanning optimization path.

Images