A multi-scene dynamic temporary road planning method and system based on GIM and remote sensing

Abstract

The application provides a multi-scene dynamic temporary road planning method and system based on GIM and remote sensing, relates to the technical field of image processing, and comprises the following steps: acquiring laser point cloud data and orthographic image data of a region to be constructed; performing data registration and fusion on the laser point cloud data, the orthographic image data and preset engineering parameters by using a preset obstacle algorithm and a spatial registration algorithm to obtain a three-dimensional geographic information dataset; analyzing key passing parameters of construction machinery and the three-dimensional geographic information dataset by using a passable domain analysis algorithm to determine passability parameters of the construction machinery; inputting the three-dimensional geographic information dataset, the passability parameters and a coordinate to be constructed into a preset path planning and optimization algorithm to perform calculation, determining an optimal mechanical approach path, classifying roads in the optimal mechanical approach path by using a spatial overlay algorithm, and obtaining corresponding road classification engineering quantities.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a multi-scene dynamic temporary road planning method and system based on GIM and remote sensing. Background Technology

[0002] Existing temporary road planning for power transmission lines mainly relies on manual surveying for route selection and design, which is greatly affected by human factors, resulting in insufficient planning accuracy and applicability. While some GIS (Geographic Information System)-based route planning methods can achieve automatic route selection, they fail to fully consider engineering constraints such as the passage conditions of construction machinery and the distribution of towers, leading to discrepancies between the planning results and the actual construction environment. If the designed road may not meet the normal passage requirements of construction machinery in terms of slope, width, and turning radius, it may cause problems such as slippage, overturning, and inability to turn during operation, affecting construction progress and safety.

[0003] With the development of remote sensing and UAV technologies, some solutions have begun to utilize remote sensing imagery and point cloud data for terrain modeling and obstacle identification, improving the accuracy of terrain data. However, these solutions mostly remain at the stage of terrain reconstruction or auxiliary analysis, and have not yet formed an automated planning process for temporary road design. Furthermore, the application of existing GIM models in power transmission projects is mainly focused on line design and asset management, lacking deep integration with remote sensing data. This makes it impossible to generate temporary road solutions that meet the passage requirements of construction equipment based on tower parameters, tower weights, and terrain features in the GIM (Grid Information Model), thus failing to meet the actual needs of power transmission project construction. Summary of the Invention

[0004] In view of this, the present invention proposes a multi-scenario dynamic temporary road planning method and system based on GIM and remote sensing.

[0005] The technical solution of this invention is implemented as follows: The first aspect of this invention provides a multi-scenario dynamic temporary road planning method based on GIM and remote sensing, comprising: Acquire laser point cloud data and orthophoto data of the area to be constructed; The laser point cloud data, orthophoto data, and preset engineering parameters are registered and fused using a preset obstacle algorithm and a spatial registration algorithm to obtain a three-dimensional geographic information dataset. The preset engineering parameters include tower coordinates extracted from the GIM model. The three-dimensional geographic information dataset includes terrain elevation, land cover, image texture, and GIM engineering attributes. The key accessibility parameters of the construction machinery and the three-dimensional geographic information dataset are analyzed by the accessibility domain analysis algorithm to determine the accessibility parameters of the construction machinery. The key accessibility parameters include terrain elevation, slope, aspect, obstacle type and density, ground bearing capacity, shortest distance to existing roads, and surface roughness. The three-dimensional geographic information dataset, the accessibility parameters, and the coordinates to be constructed are input into a preset path planning and optimization algorithm for calculation to determine the optimal machinery access path. The roads in the optimal machinery access path are classified using a spatial overlay algorithm to obtain the corresponding road classification engineering quantities.

[0006] Based on the above technical solutions, preferably, the step of using a preset obstacle algorithm and a spatial registration algorithm to perform data registration and fusion of the laser point cloud data, orthophoto data, and preset engineering parameters to obtain a three-dimensional geographic information dataset includes: The laser point cloud data is separated using a point cloth simulation filtering algorithm to obtain ground points and non-ground points; Using the ground points as a reference, the orthophoto data, the preset engineering parameters, and the digital elevation model generated based on the laser point cloud data are registered to the same coordinate system using an iterative nearest point algorithm to obtain registration data. The registered data is fused to obtain a three-dimensional geographic information dataset.

[0007] Based on the above technical solutions, preferably, the step of using a preset obstacle algorithm and a spatial registration algorithm to perform data registration and fusion of the laser point cloud data, orthophoto data, and preset engineering parameters to obtain a three-dimensional geographic information dataset further includes: Based on the spectral information of the ground points, the non-ground points, and the orthophoto data, and combined with the image segmentation algorithm, the boundaries and attributes of various obstacles in the area to be constructed are identified, and at least some of the key passage parameters are obtained.

[0008] Based on the above technical solutions, preferably, the three-dimensional geographic information dataset includes multiple sub-layers; the step of fusing the registered data to obtain the three-dimensional geographic information dataset includes: Define a uniform regular grid that covers the area to be constructed and has a fixed resolution. Resample the registered digital elevation model to the uniform regular grid to generate a digital elevation layer. The registered orthophoto is mapped onto the unified regular grid to generate a true-color image texture layer; The boundaries and attributes of the obstacles are rasterized to the unified rule grid to generate a surface cover and obstacle classification layer and an obstacle height / density layer. The registered preset engineering parameters are overlaid as vector layers and spatially associated with the unified rule grid.

[0009] Based on the above technical solutions, preferably, the step of analyzing the key accessibility parameters of the construction machinery and the three-dimensional geographic information dataset using a accessibility domain analysis algorithm to determine the accessibility parameters of the construction machinery includes: The area to be constructed is divided into multiple uniform, regular grid units on a horizontal plane; After normalizing the key access parameters within each rule grid cell, a weighted sum is performed to obtain the cell comprehensive cost, and the cell comprehensive cost is determined as the accessibility parameter of the construction machinery.

[0010] Based on the above technical solutions, preferably, the step of inputting the three-dimensional geographic information dataset, the accessibility parameters, and the coordinates to be constructed into a preset path planning and optimization algorithm for calculation to determine the optimal machinery access path includes: The improved A* algorithm is used to analyze the three-dimensional geographic information dataset, the accessibility parameters, and the coordinates to be constructed to determine the initial path node sequence; the improved A* algorithm is an A* algorithm with added penalty terms for changes in path direction. The initial path node sequence is smoothed using a B-spline curve fitting algorithm to determine the optimal mechanical entry path.

[0011] Based on the above technical solutions, preferably, the classification of roads in the optimal mechanical access path using the spatial overlay algorithm to obtain the corresponding road classification workload includes: The optimal mechanical access path is compared with the existing road vector layer in the base map data through spatial overlay algorithm analysis, and the roads are classified according to preset rules.

[0012] Furthermore, a second aspect of the present invention provides a multi-scenario dynamic temporary road planning system based on GIM and remote sensing, comprising: a data acquisition module, a registration and fusion module, a traffic analysis module, and a route determination module; wherein, The data acquisition module is configured to acquire laser point cloud data and orthophoto data of the area to be constructed. The registration and fusion module is configured to use a preset obstacle algorithm and a spatial registration algorithm to register and fuse the laser point cloud data, orthophoto data, and preset engineering parameters to obtain a three-dimensional geographic information dataset; the preset engineering parameters include pole coordinates extracted from the GIM model; the three-dimensional geographic information dataset includes terrain elevation, land cover, image texture, and GIM engineering attributes; The accessibility analysis module is configured to analyze the key accessibility parameters of the construction machinery and the three-dimensional geographic information dataset through the accessibility domain analysis algorithm to determine the accessibility parameters of the construction machinery; the key accessibility parameters include terrain elevation, slope, aspect, obstacle type and density, ground bearing capacity, shortest distance to existing roads, and surface roughness; The path determination module is configured to input the three-dimensional geographic information dataset, the accessibility parameters, and the coordinates to be constructed into a preset path planning and optimization algorithm for calculation, determine the optimal machinery access path, and classify the roads in the optimal machinery access path using a spatial overlay algorithm to obtain the corresponding road classification engineering quantities.

[0013] More preferably, a third aspect of the present invention provides an electronic device, including a processor and a memory; the memory stores a computer program, wherein the computer program, when executed by the processor, implements the multi-scenario dynamic temporary road planning method based on GIM and remote sensing described in the first aspect.

[0014] More preferably, the fourth aspect of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the multi-scenario dynamic temporary road planning method based on GIM and remote sensing described in the first aspect.

[0015] The multi-scenario dynamic temporary road planning method and system based on GIM and remote sensing of the present invention has the following advantages over the prior art: 1. By utilizing preset obstacle algorithms and spatial registration algorithms, laser point cloud data, orthophoto data, and preset engineering parameters such as tower coordinates extracted from GIM models are registered and fused to obtain a 3D geographic information dataset containing terrain elevation, surface cover, image texture, and GIM engineering attributes. This effectively solves the problem of insufficient data utilization. Combined with the accessibility domain analysis algorithm, the key access parameters of construction machinery and the 3D geographic information dataset are comprehensively analyzed to determine the accessibility parameters of construction machinery. Finally, the optimal machinery entry path and road classification engineering volume are determined, significantly improving the automation level and accuracy of path planning. This effectively overcomes the subjectivity and bias of human route selection and ensures a high degree of consistency between the planning results and actual construction needs.

[0016] 2. By dividing the construction area into uniform, regular grid cells, a unified standard framework is provided for data processing across the entire area. Each grid cell has the same size and shape, allowing key accessibility parameters at different locations to be compared and analyzed on the same spatial scale. Weighted summation assigns different weights to each parameter based on their importance to accessibility, highlighting the impact of key factors. This allows for a more scientific and accurate assessment of the accessibility of construction machinery within each grid cell, providing more accurate information for subsequent path planning and thus improving the rationality and reliability of the path planning.

[0017] 3. After determining the optimal machinery access path, the roads along the path are classified using a spatial overlay algorithm. This accurate classification of roads into different categories facilitates a more detailed understanding of road characteristics and construction requirements, providing a more precise basis for subsequent quantity calculations. By classifying roads and calculating the corresponding quantities, the required materials, manpower, and equipment resources can be more accurately estimated, enabling the development of a reasonable construction budget. Furthermore, the classified quantities allow for effective monitoring and management of construction progress and quality, timely identification and resolution of problems, ensuring the project proceeds smoothly according to plan, and improving the economic efficiency and management level of the construction project. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a multi-scenario dynamic temporary road planning method based on GIM and remote sensing provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a multi-scenario dynamic temporary road planning system based on GIM and remote sensing, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] In some embodiments, such as Figure 1 As shown, Figure 1 This is a flowchart illustrating a multi-scenario dynamic temporary road planning method based on GIM and remote sensing, provided by an embodiment of the present invention. The multi-scenario dynamic temporary road planning method based on GIM and remote sensing provided by the present invention includes: S110 acquires laser point cloud data and orthophoto data of the area to be constructed.

[0022] In this embodiment, laser point cloud data and orthophoto data of the area to be constructed can be acquired through a combination of aerial and ground-based acquisition. Aerial acquisition utilizes a flight platform equipped with a real-time dynamic differential (RTK) module, along with a simultaneously mounted lidar and aerial survey camera, to acquire laser point cloud data and high-resolution orthophoto data of the construction area. Ground-based acquisition employs a GNSS RTK surveying instrument to establish ground control points within the survey area and collect coordinates of characteristic features to verify the absolute accuracy of the aerial data. For areas obstructed by the UAV's view, a handheld 3D laser scanner can be used for refined supplementary surveying.

[0023] S120 uses a preset obstacle algorithm and a spatial registration algorithm to register and fuse laser point cloud data, orthophoto data, and preset engineering parameters to obtain a three-dimensional geographic information dataset. The preset engineering parameters include tower coordinates extracted from the GIM model. The three-dimensional geographic information dataset includes terrain elevation, land cover, image texture, and GIM engineering attributes.

[0024] Here, the preset engineering parameters can be directly read from the existing GIM of the power grid construction project. In addition to tower coordinates, they can also include engineering design parameters such as tower type, foundation structure and tower weight.

[0025] In some embodiments, a preset obstacle algorithm and a spatial registration algorithm are used to register and fuse laser point cloud data, orthophoto data, and preset engineering parameters to obtain a three-dimensional geographic information dataset, including: The laser point cloud data is separated by a point cloth simulation filtering algorithm to obtain ground points and non-ground points; Using ground points as a reference, orthophoto data, preset engineering parameters, and digital elevation models generated based on laser point cloud data are registered to the same coordinate system through an iterative nearest-point algorithm to obtain registration data. The registered data is then fused to obtain a 3D geographic information dataset.

[0026] In this embodiment, a terrain height field representation is constructed from the input laser point cloud data, and a point cloth simulation filtering algorithm is used to separate ground points from non-ground points in the laser point cloud data. By constructing a virtual cloth mesh on the point cloud height field, gravity and adjacency constraints are applied to the cloth nodes to achieve iterative updates of the cloth sagging process.

[0027] The cloth node update satisfies: ; in, and Let i be the position and velocity of node i at iteration step k, respectively.

[0028] Node speed update satisfies: ; in, It is the acceleration due to gravity. For node quality, This is the spring stiffness coefficient. Let i be the set of neighboring nodes of node i.

[0029] When the cloth node position reaches the terrain height, the following constraints apply: ; The fabric is aligned with the ground level. After the fabric is balanced, the point cloud below the fabric is marked as ground points, and the point cloud above the fabric is marked as non-ground points, thus obtaining preliminary land feature attribute labels for the point cloud. Local geometric features are extracted from the filtered non-ground point cloud to distinguish different types of obstacle structures. Based on the covariance matrix of the local neighborhood of the point cloud, eigenvalues ​​are solved. Obtain the flatness parameters.

[0030] Regarding the calculation of the local covariance matrix, for any point in the point cloud... Using it as the center, select a spherical neighborhood of radius R or the k nearest points to form a local point set. The coordinates of each point are .

[0031] Calculate the centroid (mean) of the neighborhood points: ; The 3×3 covariance matrix C of this local point set is defined as: ; in, It is a 3×1 column vector. Indicates transpose, therefore It is a 3×3 matrix. Summing and then dividing by the number of points... This yields a symmetric positive semi-definite matrix. .

[0032] For covariance matrix Perform eigenvalue decomposition: ; Three eigenvalues ​​were obtained and its corresponding eigenvectors Eigenvectors constitute the principal directions of the local coordinate system, while the magnitude of the eigenvalues ​​characterizes the variance of the point set along each principal direction.

[0033] Based on eigenvalues, the following geometric descriptors can be defined: Flatness: Used to measure how close a local shape is to a plane.

[0034] ; Linearity: Used to measure how close a local shape is to a line.

[0035] ; Scattering degree: Used to measure the uniformity of the distribution of local shape in all directions.

[0036] ; Simultaneously, roughness characteristics are calculated based on the residual from the point to the local plane: ; in, The coordinates of the neighboring points, The mean of the neighborhood points. This represents the local plane normal vector. Geometric features (flatness, linearity, scattering, and roughness) together constitute a multi-dimensional discrimination criterion, capable of effectively distinguishing different types of obstacles.

[0037] Considering that buildings typically exhibit high flatness (Planarity≈1) and low roughness (Roughness≈0), and low scattering ( Smaller). Trees or shrubs typically exhibit low flatness and high scattering ( Larger diameters and high roughness, with isotropic geometry and irregular surfaces. Rod-shaped structures (such as utility poles and towers) primarily exhibit high linearity (…). Meanwhile, the flatness and roughness are both low, and the geometric shape exhibits a clear directional distribution. By setting thresholds or rules corresponding to the above feature combinations, automated preliminary identification and differentiation of different types of obstacles can be achieved, providing key inputs for the differential cost calculation in subsequent accessibility analysis.

[0038] Based on the exterior orientation parameters of the point cloud and the image, the 3D point cloud coordinates are projected onto the corresponding pixel positions in the orthophoto, thus establishing the point-to-pixel correspondence. The projection relationship satisfies:

[0039] ; Through this mapping, the point cloud can obtain the spectral values ​​of the corresponding pixels and image category information.

[0040] Perform spectral feature analysis and automatic segmentation on orthophotos. Taking vegetation identification as an example, calculate the normalized vegetation index:

[0041] ; when When the value exceeds a preset threshold, the corresponding pixel is marked as a vegetation area. Furthermore, image pixels can be classified using an energy-minimization-based image segmentation model, the optimization objective of which is expressed as:

[0042] ; in, The total energy (cost function) of the segmentation result. For pixels The assigned category labels (such as vegetation, buildings, etc.). For data items, representing pixels Marked as category The cost reflects the degree of matching between pixel spectral features and category. For smoothing terms, it represents adjacent pixels. and The penalty for assigning different labels is used to maintain spatial continuity. It is a neighborhood system for pixels (usually 4-neighbor or 8-neighbor). The balancing parameter is used to adjust the weights between the data items and the smoothing term. The specific process of model solving and classification result generation is as follows: Initialization and cost calculation: First, define the target category set, such as... For each pixel in the image Based on its multi-band spectral values, its classification is calculated. Matching cost for each category This forms the cost vector for that pixel. The entire image is treated as a graph, with each pixel as a node and its neighborhood as edges. The cost vector is then substituted into the data item. and combined with smoothing terms This constitutes a specific global energy function. The graph cut algorithm is used to minimize the energy function described above. This algorithm iteratively executes... The "swap" operation continuously attempts to adjust the pixel labels. In each swap, the algorithm precisely calculates; if a pixel belongs to... and Will redistributing labels to a group of pixels from both categories reduce the total energy? By traversing all class pairs and iterating until energy convergence, a value that minimizes the total energy is eventually found. Minimal or near-minimum label configuration .

[0043] The output of the above optimization process It is a label matrix with the same size as the input image. Each element (corresponding to a cell) in this matrix stores an integer label value, which directly corresponds to a preset set of categories. The system assigns a label matrix to each pixel (e.g., 1 for bare land, 2 for vegetation, 3 for buildings, and 4 for water). This label matrix is ​​then rendered, resulting in a classification map where each pixel has a clearly defined category attribute. Through this process, the model automatically assigns an optimal category label to each pixel, resulting in a pixel-level classification of multiple categories, including bare land, vegetation, buildings, and water. This result is not generated by setting multiple isolated threshold ranges, but rather through global energy optimization, automatically generated while comprehensively considering the spectral attributes of each pixel and the spatial consistency with surrounding pixels. By solving the above model, pixel classification results including categories such as bare land, vegetation, buildings, and water can be obtained.

[0044] Based on the initial labels of the point cloud, image segmentation categories, and local geometric features of the point cloud, obstacle category identification is performed for each point cloud point using a fusion discrimination rule. The fusion classification satisfies:

[0045] ; in, Point cloud ground / non-ground labels; Image feature type; It includes geometric features such as height, flatness, and roughness.

[0046] Through the above fusion process, obstacles such as trees, buildings, water bodies, and small structures in point clouds can be accurately classified.

[0047] For the classified point cloud, based on the spatial distribution of points of each category, through... An algorithm or convex hull algorithm is used to construct the obstacle boundary. The boundary construction satisfies:

[0048] ; in, For category The collection of point clouds, This represents the boundary polygon of the corresponding obstacle. Further extraction of the obstacle's height, area, and density features is performed, and the results are rasterized to a unified coordinate system to provide an obstacle attribute raster for subsequent traversable domain calculations.

[0049] ; in, For category Obstacle type labels.

[0050] Using ground control points as a reference, spatial registration algorithms such as Iterative Closest Point (ICP) are employed to accurately register the high-precision digital elevation model (DEM) generated from laser point clouds, orthophoto data, and tower coordinates extracted from the GIM model to the same coordinate system. Then, through data fusion, a high-precision 3D geographic information dataset integrating terrain elevation, land cover, image texture, and GIM engineering attributes is generated. Specifically, this may include the following steps:

[0051] Establishing an absolute coordinate datum. Determine a unified plane coordinate system and elevation datum for the project. Use the coordinates of ground control points measured in the field using GNSS RTK as the true reference values ​​with absolute accuracy.

[0052] Absolute positioning based on remote sensing data. Using ground points, perform joint or separate absolute geolocation on image data and laser point cloud data acquired by UAV aerial surveys.

[0053] For orthophotos: During photogrammetric processing, the coordinates of ground control points are used as strong constraints in the bundle adjustment algorithm to optimize the exterior orientation elements of the image, ensuring that the generated orthophoto has accurate absolute geographic coordinates. The adjustment process is optimized based on a collinearity condition equation, which is as follows:

[0054] ; ; in, For image point coordinates, The ground coordinates of the corresponding ground control points. For camera focal length, Let the principal point coordinates be... Coordinates of the photography center Rotation matrix Element.

[0055] For laser point clouds: By identifying feature points in the point cloud that correspond to ground control points, a three-dimensional similarity transformation parameter is calculated to transform the entire point cloud to the target coordinate system. The three-dimensional similarity transformation model is as follows:

[0056] ; in, For the target coordinate system coordinates, The source point cloud coordinates, As a scale factor, To pass through the rotation angle Defined rotation matrix, The above parameters are calculated using the least squares method to achieve absolute positioning of the point cloud and further generate a high-precision digital elevation model with absolute coordinates.

[0057] Based on absolute positioning, an iterative nearest-point algorithm is used to perform relative fine registration of the already registered digital elevation model, orthophoto, and GIM coordinates, eliminating residual systematic biases. Stable feature lines for building edges and road contours are extracted between the digital elevation model and the orthophoto, forming source point sets respectively. With target point set .

[0058] The core of the iterative closest point algorithm is to iteratively solve for an optimal rigid body transformation. , so that the transformed and The error between them is minimized. Its objective function is:

[0059] ; in, yes Zhongyu The nearest neighbor is found. The optimal solution for each iteration is obtained using singular value decomposition. and The process continues until the transformation converges, achieving pixel-level alignment between the digital elevation model (GIM) and the orthophoto. For the tower design coordinates in the GIM model, they are matched with the actual target tower locations identified on the orthophoto / digital elevation model. Similarly, a three-dimensional similarity transformation is calculated to register the entire GIM model to the real geographic space, ensuring consistency between the design coordinates and the actual location.

[0060] In some embodiments, laser point cloud data, orthophoto data, and preset engineering parameters are registered and fused using a preset obstacle algorithm and a spatial registration algorithm to obtain a three-dimensional geographic information dataset, which further includes: Based on the spectral information of ground points, non-ground points, and orthophoto data, and combined with image segmentation algorithms, the boundaries and attributes of various obstacles in the area to be constructed are identified, and at least some key passage parameters are obtained.

[0061] In some embodiments, the 3D geographic information dataset includes multiple sub-layers; the registration data is fused to obtain the 3D geographic information dataset, including: Define a uniform regular grid that covers the area to be constructed and has a fixed resolution. Resample the registered digital elevation model to the uniform regular grid to generate a digital elevation layer. The registered orthophoto is mapped onto a uniform regular grid to generate a true-color image texture layer; The boundaries and attributes of obstacles are rasterized into a uniform regular grid to generate a surface cover and obstacle classification layer and an obstacle height / density layer. The registered preset engineering parameters are overlaid as vector layers and spatially associated with a unified rule grid.

[0062] Here, the data structure of the three-dimensional geographic information dataset can be characterized as follows: .

[0063] S130 uses a passability domain analysis algorithm to analyze the key passability parameters of construction machinery and a three-dimensional geographic information dataset to determine the passability parameters of the construction machinery. The key passability parameters include terrain elevation, slope, aspect, obstacle type and density, ground bearing capacity, shortest distance to existing roads, and surface roughness.

[0064] In some embodiments, the key accessibility parameters of the construction machinery and the three-dimensional geographic information dataset are analyzed using a accessibility domain analysis algorithm to determine the accessibility parameters of the construction machinery, including: The area to be constructed is divided into multiple uniform, regular grid units on a horizontal plane; After normalizing the key access parameters within each regular grid cell, a weighted sum is performed to obtain the cell's comprehensive cost, which is then determined as the accessibility parameter for the construction machinery.

[0065] In this embodiment, the study area is divided into sections of size [size missing] on the horizontal plane. The data is divided into regular grid cells. For each cell, terrain elevation, slope, aspect, obstacle type and density, ground bearing capacity, shortest distance to existing roads, and surface roughness are extracted from the fused data. The extracted key access parameters are used as sub-factors to constitute the cost, normalized, and then weighted and summed to obtain the comprehensive cost of the cell, which is used to determine the accessibility of construction machinery.

[0066] For example, the following basic attributes are extracted from the fused data for each cell: terrain elevation. ,slope (Or use slope percentage), aspect Obstacle types and densities (e.g., 0: barrier-free, 1: sparse vegetation, 2: dense woodland, 3: buildings / water bodies), ground bearing capacity (or soil type), shortest distance to existing roads Surface roughness Etc. Define several sub-factors used to constitute the cost: slope factor. Slope aspect / facing factor Obstacles Carrying factor Distance factor Width / Turn Feasibility Factor Environmental limiting factors Etc. Normalize each sub-factor to ensure its value range is... The larger the value, the more unfavorable it is for passage (i.e., the higher the cost of passage).

[0067] The slope factor can be expressed as: ; The barrier factor can be expressed as: ; in, For the preset obstacle weight table, the points in the parentheses of the formula It is a parameter placeholder, for example: the default value for sparse vegetation is 0.3, the default value for dense woodland is 0.7, and the default value for buildings / water bodies is 1.0.

[0068] The carrying factor can be expressed as: ; in, For the minimum required load-bearing capacity of machinery, if ,but Otherwise, it will be close to 0.

[0069] The distance factor can be expressed as: ; in, This is the distance threshold.

[0070] To take into account the minimum turning radius of construction machinery With minimum road width Morphological image processing methods are employed. First, absolute obstacles (such as...) are processed. The expansion operation is performed on the region, with an expansion radius of [missing information]. or The set of grid cells covered after expansion. Marked as a geometrically infeasible region. If the mesh... If it is unreachable, it is directly marked as unreachable; otherwise, its geometric feasibility factor contribution is 0.

[0071] The unit comprehensive cost is obtained by weighted summation of the normalized sub-factors. : ; in, For the first Sub-factors (e.g.) ), For the corresponding weights, satisfying ; This is a smoothing / curvature penalty term used to reduce path turns and improve road smoothness; The weights are for the smoothing term.

[0072] The smoothing term can be approximated using neighborhood second-order differences or curvature: ; For grids that satisfy any "hard constraints," such as those falling on buildings / water bodies, exceeding maximum slope, having insufficient bearing capacity, or being covered after geometric expansion, their cost is set to... During path search, certain grid cells are treated as obstacles. The remaining grid cells are subject to soft constraints based on cost, allowing path crossings but at a high cost. The weight vector is adjusted according to the construction scenario, such as rainy season, nighttime, heavy machinery, or ecological protection zones. With threshold parameters, such as It also allows enabling / disabling certain areas, such as temporary restricted areas. Scene adjustment parameters can be set via configuration files or through the interface.

[0073] S140: Input the three-dimensional geographic information dataset, accessibility parameters, and coordinates to be constructed into the preset path planning and optimization algorithm for calculation, determine the optimal machinery access path, and classify the roads in the optimal machinery access path using the spatial overlay algorithm to obtain the corresponding road classification engineering quantities.

[0074] In some embodiments, the three-dimensional geographic information dataset, accessibility parameters, and coordinates to be constructed are input into a preset path planning and optimization algorithm for calculation to determine the optimal machinery access path, including: The improved A* algorithm is used to analyze the 3D geographic information dataset, accessibility parameters, and coordinates to be constructed to determine the initial path node sequence. The improved A* algorithm is an A* algorithm with added penalty terms for changes in path direction. The initial path node sequence is smoothed using a B-spline curve fitting algorithm to determine the optimal mechanical entry path.

[0075] In this embodiment, with the starting point S as the initial node and the target point G as the ending node, the evaluation function for each node n of the initial and ending nodes is: ; in, It is the cumulative actual cost from the starting point S to node n. It is a heuristic estimate from n to the target G, obtained by iteratively expanding the open list. The smallest node is used until the target point G is reached. Then, the initial path node sequence is obtained by backtracking the parent node. The initial path node sequence is used as the control points of the B-spline curve. A smooth and continuous road centerline is generated using the B-spline curve model. By adjusting the curve parameters and control points, it is ensured that the radius of curvature of the fitted path is not less than the minimum turning radius required by the construction machinery throughout the entire process. This ensures that the optimal machinery access path has geometrically continuous curvature, allowing the construction machinery to enter the site directly.

[0076] For example, extracting node sequences from the path output by the A* algorithm. ,in, These are two-dimensional planar coordinates. Since Algorithm A is based on a grid, the path has jagged turns and requires preprocessing.

[0077] For three consecutive approximately collinear nodes Calculate vector and The included angle ,like (For example If the angle is 5°, then remove the intermediate nodes. Along the path at a fixed arc length Resampling yields a new node sequence. , as the candidate set of initial control points for the B-spline curve.

[0078] A cubic uniform B-spline curve is used for fitting, which strikes a balance between continuity and computational efficiency. Given control points... and node vectors (Usually a uniform distribution is assumed, i.e.) Define B-spline curve :

[0079] ; in, The basis functions are cubic B-spline functions, calculated using the de Boor recurrence relation: ; ; To ensure that the curve's start and end points are respectively aligned with the path's start point. and the end point To achieve overlap, the endpoint repetition technique is used, which involves repeating the first and last nodes in the node vector. (here) ) times, and order , .

[0080] Minimum turning radius of construction machinery Requires the radius of curvature at any point on the path. For parametric curves curvature The calculation formula is:

[0081] ; radius of curvature To ensure constraints, a set of parameter values ​​is sampled uniformly along the curve. Calculate the corresponding radius of curvature If it exists Then the control points of the B-spline curve need to be adjusted. The adjustment process is based on the gradient information of the curvature at the control points:

[0082] ; in, .

[0083] The control points are iteratively adjusted using gradient descent or optimization-based methods, with the optimization objective being: ; In the formula, As the initial control point, This is the regularization coefficient, used to prevent excessive deformation. Through constraint optimization, the adjusted B-spline curve satisfies curvature constraints.

[0084] B-spline curves that satisfy curvature constraints Arc length intervals as required by construction machinery navigation Discretization yields the final smooth path point sequence. It also includes the tangent direction at each point. and curvature The information is provided to the mechanical control system. Through the above steps, the transformation from a discrete grid path to a continuous smooth path that satisfies mechanical kinematic constraints is achieved, enabling the path to be directly used for precise navigation of construction machinery.

[0085] In some embodiments, a spatial overlay algorithm is used to classify roads in the optimal mechanical access path to obtain the corresponding road classification workload, including: The optimal mechanical access path is compared with the existing road vector layer in the base map data through spatial overlay algorithm analysis, and the roads are classified according to preset rules.

[0086] In this embodiment, spatial overlay analysis is used to compare the machinery access path with the existing road vector layer in the base map data, and the road construction type is automatically classified according to preset rules. The map mapping module is then invoked to perform visualization rendering based on the classified road construction types using different colors, and the lengths of the classified roads are marked before being output.

[0087] For example, first prepare the centerline (vector line) of the temporary road generated by the plan, as well as the existing road layer in the base map data, which is also a vector line with attributes such as width. Compare the spatial position of each segment of the machinery access path with the existing road. The core is to determine if there is an "overlapping" relationship. This "overlapping" does not require perfect overlap; it is usually achieved by creating a "buffer zone" of a certain width for the existing road. Mark "Utilizing Road": If a planned path segment falls entirely within the "buffer zone" of an existing road, and the attributes of the existing road, such as width and material, meet the requirements for construction machinery access, the system marks this path segment as "Utilizing Road". Mark "Expanding Road": If a planned path segment overlaps with an existing road, but the existing road's current attributes, such as insufficient width, do not meet the construction requirements, the system marks this segment as "Expanding Road," meaning that the existing road needs to be widened or reinforced. Mark "New Road": For road segments in the planned path that do not overlap with any existing road, the system automatically marks them as "New Road," indicating that a completely new road needs to be built.

[0088] Based on this, the system uses different colors to visualize and render the roads according to their classified construction types, and marks the lengths of the classified roads before outputting the results. The process for generating the master plan and statistical tables is as follows: 1. Generate the master plan: Based on the previously generated and categorized planning path data, the system uses the map drawing module for visualization and rendering. "New Roads," "Expanded Roads," and "Utilized Roads" are drawn on the same geographic base map using three different colors, such as pink, green, and blue. 2. Automatically calculate road lengths: The system iterates through all classified path segments and automatically calculates the geometric length of each segment using its vector coordinate information. Since the data has been registered to the real coordinate system, the precise metric lengths of each road segment and the total road length can be directly obtained. 3. Associate attributes and generate statistical tables: The system associates the "length" information of each road segment with its "construction type" attribute and summarizes the data. Subsequently, the program automatically fills these data into a preset template table, instantly generating a categorized engineering quantity statistical table that clearly lists the length and proportion of each type of road. 4. Output the final results. Finally, integrate the overall planning map and the classified engineering quantity statistics table into a complete planning results report, which shows the route planning list for the entire line.

[0089] In some embodiments, please refer to Figure 2 , Figure 2 This is a schematic diagram of a multi-scenario dynamic temporary road planning system based on GIM and remote sensing, provided by an embodiment of the present invention. The present invention provides a multi-scenario dynamic temporary road planning system 200 based on GIM and remote sensing, comprising: a data acquisition module 210, a registration and fusion module 220, a traffic analysis module 230, and a route determination module 240; wherein,

[0090] Data acquisition module 210 is configured to acquire laser point cloud data and orthophoto data of the area to be constructed; The registration and fusion module 220 is configured to use a preset obstacle algorithm and a spatial registration algorithm to register and fuse laser point cloud data, orthophoto data and preset engineering parameters to obtain a three-dimensional geographic information dataset; the preset engineering parameters include tower coordinates extracted from the GIM model; the three-dimensional geographic information dataset includes terrain elevation, land cover, image texture and GIM engineering attributes; The accessibility analysis module 230 is configured to analyze the key accessibility parameters of the construction machinery and the three-dimensional geographic information dataset through the accessibility domain analysis algorithm to determine the accessibility parameters of the construction machinery. The key accessibility parameters include terrain elevation, slope, aspect, obstacle type and density, ground bearing capacity, shortest distance to existing roads, and surface roughness. The path determination module 240 is configured to input the three-dimensional geographic information dataset, accessibility parameters and coordinates to be constructed into a preset path planning and optimization algorithm for calculation, determine the optimal machinery access path, and classify the roads in the optimal machinery access path by combining the spatial overlay algorithm to obtain the corresponding road classification engineering quantities.

[0091] In some embodiments, the registration and fusion module 220 is specifically configured as follows: The laser point cloud data is separated by a point cloth simulation filtering algorithm to obtain ground points and non-ground points; Using ground points as a reference, orthophoto data, preset engineering parameters, and digital elevation models generated based on laser point cloud data are registered to the same coordinate system through an iterative nearest-point algorithm to obtain registration data. The registered data is then fused to obtain a 3D geographic information dataset.

[0092] In some embodiments, the registration and fusion module 220 is further configured as follows: Based on the spectral information of ground points, non-ground points, and orthophoto data, and combined with image segmentation algorithms, the boundaries and attributes of various obstacles in the area to be constructed are identified, and at least some key passage parameters are obtained.

[0093] In some embodiments, the three-dimensional geographic information dataset includes multiple sub-layers; the registration and fusion module 220 is further configured to: Define a uniform regular grid that covers the area to be constructed and has a fixed resolution. Resample the registered digital elevation model to the uniform regular grid to generate a digital elevation layer. The registered orthophoto is mapped onto a uniform regular grid to generate a true-color image texture layer; The boundaries and attributes of obstacles are rasterized into a uniform regular grid to generate a surface cover and obstacle classification layer and an obstacle height / density layer. The registered preset engineering parameters are overlaid as vector layers and spatially associated with a unified rule grid.

[0094] In some embodiments, the access analysis module 230 is specifically configured as follows: The area to be constructed is divided into multiple uniform, regular grid units on a horizontal plane; After normalizing the key access parameters within each regular grid cell, a weighted sum is performed to obtain the cell's comprehensive cost, which is then determined as the accessibility parameter for the construction machinery.

[0095] In some embodiments, the path determination module 240 is specifically configured as follows: The improved A* algorithm is used to analyze the 3D geographic information dataset, accessibility parameters, and coordinates to be constructed to determine the initial path node sequence. The improved A* algorithm is an A* algorithm with added penalty terms for changes in path direction. The initial path node sequence is smoothed using a B-spline curve fitting algorithm to determine the optimal mechanical entry path.

[0096] In some embodiments, the path determination module 240 is further configured to: The optimal mechanical access path is compared with the existing road vector layer in the base map data through spatial overlay algorithm analysis, and the roads are classified according to preset rules.

[0097] It should be noted that the multi-scenario dynamic temporary road planning system based on GIM and remote sensing provided in this application embodiment is based on the same application concept as the multi-scenario dynamic temporary road planning method based on GIM and remote sensing provided in this application embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned multi-scenario dynamic temporary road planning method based on GIM and remote sensing, and the repeated parts will not be described again.

[0098] In some embodiments, please refer to Figure 3 , Figure 3This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 300 provided in this embodiment includes a processor 310 and a memory 320; the memory 320 stores a computer program, wherein the computer program, when executed by the processor, implements the aforementioned multi-scenario dynamic temporary road planning method based on GIM and remote sensing.

[0099] Specifically, processor 310 may include, for example, a general-purpose microprocessor, an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. Processor 310 may also include onboard memory for caching purposes. Processor 310 may be a single processing unit or multiple processing units for performing different actions of the method flow according to embodiments of this application.

[0100] The memory 320 may be any medium capable of containing, storing, transmitting, propagating, or transmitting instructions. For example, the memory 320 may include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, apparatuses, or propagation media. Specific examples of the memory 320 include: magnetic storage devices such as magnetic tape or hard disk drives (HDDs); optical storage devices such as optical discs (CD-ROMs); and may also be random access memory (RAM) or flash memory; and / or wired / wireless communication links.

[0101] This application also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, this program implements the aforementioned multi-scenario dynamic temporary road planning method based on GIM and remote sensing. This computer-readable medium may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into that device / apparatus / system. The aforementioned computer-readable medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0102] According to embodiments of this application, a computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wired, optical fiber, radio frequency signals, etc., or any suitable combination thereof.

[0103] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments and / or claims of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application. Therefore, the scope of this application should not be limited to the above embodiments, but should be defined not only by the appended claims, but also by their equivalents. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the protection scope of this invention.

Claims

1. A multi-scenario dynamic temporary road planning method based on GIM and remote sensing, characterized in that, include: Acquire laser point cloud data and orthophoto data of the area to be constructed; The laser point cloud data, orthophoto data, and preset engineering parameters are registered and fused using a preset obstacle algorithm and a spatial registration algorithm to obtain a three-dimensional geographic information dataset. The preset engineering parameters include tower coordinates extracted from the GIM model. The three-dimensional geographic information dataset includes terrain elevation, land cover, image texture, and GIM engineering attributes. The key accessibility parameters of the construction machinery and the three-dimensional geographic information dataset are analyzed by the accessibility domain analysis algorithm to determine the accessibility parameters of the construction machinery. The key accessibility parameters include terrain elevation, slope, aspect, obstacle type and density, ground bearing capacity, shortest distance to existing roads, and surface roughness. The three-dimensional geographic information dataset, the accessibility parameters, and the coordinates to be constructed are input into a preset path planning and optimization algorithm for calculation to determine the optimal machinery access path. The roads in the optimal machinery access path are classified using a spatial overlay algorithm to obtain the corresponding road classification engineering quantities.

2. The multi-scenario dynamic temporary road planning method based on GIM and remote sensing as described in claim 1, characterized in that, The laser point cloud data, orthophoto data, and preset engineering parameters are registered and fused using a preset obstacle algorithm and a spatial registration algorithm to obtain a three-dimensional geographic information dataset, including: The laser point cloud data is separated using a point cloth simulation filtering algorithm to obtain ground points and non-ground points; Using the ground points as a reference, the orthophoto data, the preset engineering parameters, and the digital elevation model generated based on the laser point cloud data are registered to the same coordinate system using an iterative nearest point algorithm to obtain registration data. The registered data is then fused to obtain a three-dimensional geographic information dataset.

3. The multi-scenario dynamic temporary road planning method based on GIM and remote sensing as described in claim 2, characterized in that, The step of registering and fusing the laser point cloud data, orthophoto data, and preset engineering parameters using a preset obstacle algorithm and a spatial registration algorithm to obtain a three-dimensional geographic information dataset also includes: Based on the spectral information of the ground points, the non-ground points, and the orthophoto data, and combined with the image segmentation algorithm, the boundaries and attributes of various obstacles in the area to be constructed are identified, and at least some of the key passage parameters are obtained.

4. The multi-scenario dynamic temporary road planning method based on GIM and remote sensing as described in claim 3, characterized in that, The three-dimensional geographic information dataset includes multiple sub-layers; the process of fusing the registered data to obtain the three-dimensional geographic information dataset includes: Define a uniform regular grid that covers the area to be constructed and has a fixed resolution. Resample the registered digital elevation model to the uniform regular grid to generate a digital elevation layer. The registered orthophoto is mapped onto the unified regular grid to generate a true-color image texture layer; The boundaries and attributes of the obstacles are rasterized to the unified rule grid to generate a surface cover and obstacle classification layer and an obstacle height / density layer. The registered preset engineering parameters are overlaid as vector layers and spatially associated with the unified rule grid.

5. The multi-scenario dynamic temporary road planning method based on GIM and remote sensing as described in claim 1, characterized in that, The process of analyzing key access parameters of the construction machinery and the three-dimensional geographic information dataset using a accessibility domain analysis algorithm to determine the accessibility parameters of the construction machinery includes: The area to be constructed is divided into multiple uniform, regular grid units on a horizontal plane; After normalizing the key access parameters within each rule grid cell, a weighted sum is performed to obtain the cell comprehensive cost, and the cell comprehensive cost is determined as the accessibility parameter of the construction machinery.

6. The multi-scenario dynamic temporary road planning method based on GIM and remote sensing as described in claim 1, characterized in that, The step of inputting the three-dimensional geographic information dataset, the accessibility parameters, and the coordinates to be constructed into a preset path planning and optimization algorithm for calculation to determine the optimal machinery access path includes: The improved A* algorithm is used to analyze the three-dimensional geographic information dataset, the accessibility parameters, and the coordinates to be constructed to determine the initial path node sequence; the improved A* algorithm is an A* algorithm with added penalty terms for changes in path direction. The initial path node sequence is smoothed using a B-spline curve fitting algorithm to determine the optimal mechanical entry path.

7. The multi-scenario dynamic temporary road planning method based on GIM and remote sensing as described in claim 6, characterized in that, The combined spatial overlay algorithm is used to classify the roads in the optimal mechanical access path to obtain the corresponding road classification workload, including: The optimal mechanical access path is compared with the existing road vector layer in the base map data through spatial overlay algorithm, and the roads are classified according to preset rules.

8. A multi-scenario dynamic temporary road planning system based on GIM and remote sensing, characterized in that, include: The module comprises a data acquisition module, a registration and fusion module, a traffic analysis module, and a path determination module; among which, The data acquisition module is configured to acquire laser point cloud data and orthophoto data of the area to be constructed. The registration and fusion module is configured to use a preset obstacle algorithm and a spatial registration algorithm to register and fuse the laser point cloud data, orthophoto data, and preset engineering parameters to obtain a three-dimensional geographic information dataset; the preset engineering parameters include pole coordinates extracted from the GIM model; the three-dimensional geographic information dataset includes terrain elevation, land cover, image texture, and GIM engineering attributes; The accessibility analysis module is configured to analyze the key accessibility parameters of the construction machinery and the three-dimensional geographic information dataset through the accessibility domain analysis algorithm to determine the accessibility parameters of the construction machinery; the key accessibility parameters include terrain elevation, slope, aspect, obstacle type and density, ground bearing capacity, shortest distance to existing roads, and surface roughness; The path determination module is configured to input the three-dimensional geographic information dataset, the accessibility parameters, and the coordinates to be constructed into a preset path planning and optimization algorithm for calculation, determine the optimal machinery access path, and classify the roads in the optimal machinery access path using a spatial overlay algorithm to obtain the corresponding road classification engineering quantities.

9. An electronic device comprising a processor and a memory; said memory storing a computer program, wherein, When the computer program is executed by the processor, it implements the multi-scenario dynamic temporary road planning method based on GIM and remote sensing as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that, It stores a computer program, wherein the computer program, when executed by a processor, implements the multi-scenario dynamic temporary road planning method based on GIM and remote sensing as described in any one of claims 1 to 7.

Images