Intelligent optical cable laying method based on virtual reality

Abstract

The invention relates to the technical field of optical cable design analysis, and discloses an intelligent optical cable laying method based on virtual reality, which comprises the following steps: acquiring landform data of a target area, judging whether the landform of the target area is a soft soil zone, generating an optical cable laying scheme of a non-soft soil zone, and generating a soil data processing and scheme of the soft soil zone. According to the intelligent optical cable laying method based on virtual reality, the topographic condition is judged, if soft soil does not belong to the topographic condition, a proper laying scheme is automatically generated and selected according to historical experience, and if soft soil belongs to the topographic condition, data collection is carried out on a target area again. The method comprises the steps of determining whether data of this time has a large change compared with historical data, if yes, determining whether a target area meets the requirements of optical cable laying, automatically generating and selecting a proper laying scheme, and collecting data and adjusting the scheme in real time in the laying process, thereby avoiding accidents and prolonging the service life of the optical cable.

Description

Technical Field

[0001] The present invention relates to the technical field of optical cable design and analysis, and in particular to an optical cable intelligent laying method based on virtual reality. Background Art

[0002] With the rapid development of communication networks, the complexity and precision requirements of optical cable laying projects are constantly increasing. Traditional laying methods usually rely on manual operations and plan drawings, which have problems such as insufficient spatial perception, high operational risks, and suboptimal route planning. The application of virtual reality (VR) technology can provide immersive three-dimensional spatial perception, helping construction personnel to simulate the laying process in advance, optimize route design, and enhance the accuracy and safety of on-site operations. The intelligent optical cable laying method based on virtual reality establishes a three-dimensional virtual environment, digitally simulates the on-site geographic information, pipeline layout, construction equipment, etc., and combines intelligent navigation with path optimization algorithms to achieve an organic combination of virtual guidance, simulation verification, and on-site interaction.

[0003] The existing intelligent optical cable laying method based on virtual reality cannot perform topographic analysis on the target area, determine whether the topography of the target area belongs to a soft soil zone, and cannot judge the changes in data according to the topographic conditions, especially the soft soil zone, and determine whether it is necessary to re-determine the data collection point. For soft soil zones, it is impossible to determine whether there are areas in the area that can lay optical cables and meet the requirements for laying optical cables. It is easy to cause certain damage to the optical cables during the laying process, and it cannot meet the optical cable laying needs in different scenarios. It increases construction problems and risks caused by unreasonable plans, cannot reasonably allocate and utilize construction resources, reduces resource utilization efficiency, and increases costs. Its practicality has certain limitations. Summary of the Invention

[0004] The present invention provides a virtual reality-based intelligent optical cable laying method, which is used to promote the solution of the problems raised in the background technology.

[0005] The present invention provides the following technical solution: a virtual reality-based intelligent optical cable laying method, comprising: acquiring geographic data within a target area; if a collection point needs to be regenerated, setting regional data for the target area;

[0006] At each sub-collection point At , soil data is collected to obtain a soil data set:

[0007] D ijk ={d ijk1 , d ijk2 ,...,d ijkm}; Among them, D ijkt represents the t-th soil data indicator;

[0008] Set the soil data index threshold that meets the requirements: T th ={T th1 , T th2 ,...,T thm}; where T tht represents the threshold value of the t-th soil data indicator;

[0009] For each sub-collection point Define a soil judgment function to determine whether the soil data meets the requirements:

[0010] Count each sub-region S i The number of sub-collection points that meet the requirements:

[0011] Set a paving judgment function to judge the sub-area S i Whether it meets the laying requirements:

[0012] Among them, α is the required ratio threshold, β is the depth standard deviation threshold, and S_D_depth is the depth standard deviation;

[0013] If FCT i = True, then the sub-area is determined to meet the paving requirements; if FCT i = False, then the sub-area is judged to be not in compliance with the paving requirements;

[0014] Count the number of sub-areas that meet the paving requirements:

[0015] Determine and generate optical cable laying plan.

[0016] As an optional solution of the optical cable intelligent laying method based on virtual reality of the present invention, wherein: setting the target area data, specifically: obtaining the target area range R; obtaining the elevation data e; extracting the dimension of the elevation data e, recorded as em×en;

[0017] Get the latitude range of the target area [lat min ,lat max ] and longitude range [lon min ,lon max ];

[0018] Calculate the latitude interval between two adjacent elevation data points in the latitude direction:

[0019] Calculate the longitude interval between two adjacent elevation data points in the longitude direction:

[0020] For each grid point (i, j) in the elevation data e, calculate the partial derivatives of the elevation in the latitude and longitude directions:

[0021] Compute terrain gradients based on partial derivatives:

[0022] Combine the latitude and longitude and gradient values ​​of each point into a feature vector: X = [lat, lon, terrain_gradient];

[0023] The feature vectors of all points form a data set Df: Df = {x1, x2, ..., x m};

[0024] Set the number of clusters n;

[0025] Randomly select the first cluster center c1: c1 = x i ;

[0026] For each data point x j , calculate its distance to the nearest cluster center: d j =min 1≤i≤k ||x j -ci||;

[0027] The probability of selecting the next cluster center ck is:

[0028] For each data point x j , calculate its distance to all cluster centers ci||x j -ci||;

[0029] x j Assign to the closest cluster center: cclosest = argmin ci ||x j -ci||;

[0030] For each cluster k, the new cluster center ck new is the mean of all data points in the cluster:

[0031] Among them, C k are all data points in cluster k, |C k | is the number of data points in cluster k;

[0032] For each cluster C k , extract the geographic coordinates of all data points in the cluster:

[0033] C k ={(lat k，1 ,lon k，1), (lat k，2 ,lon k，2 ),...,(lat k，nk ,lon k，nk )}; where nk is cluster k

[0034] The number of data points in ;

[0035] For each cluster C k , calculate its bounds:

[0036]

[0037] Let the boundary of each cluster be represented as a rectangular region:

[0038] S k =(lat min，k ,lon min，k ,lat max，k ,lon max，k );

[0039] Integrate all rectangular areas to form a set of sub-areas: S k ={S1, S2, ..., S n};

[0040] For each sub-region, m collection points are randomly generated inside it, and the collection point set is expressed as:

[0041] P i ={P i1 , P i2 ,...,P im}; Among them, P ij =(lat ij ,lon ij );

[0042] For each collection point P ij , randomly generate k sub-collection points with different depths, then the sub-collection point set is expressed as: in,

[0043] As an optional solution of the optical cable intelligent laying method based on virtual reality of the present invention, wherein: judging and generating the optical cable laying plan, specifically: extracting all FCT i = True, forming a qualified sub-region set;

[0044] Extract any two sub-regions from the qualified sub-region set and define them as the first sub-region and the second sub-region respectively; extract the boundary lines of the first sub-region and the second sub-region respectively and define them as the first boundary line and the second boundary line;

[0045] Generating a plurality of boundary points on the first boundary line to form a first boundary point set; generating a plurality of boundary points on the second boundary line to form a second boundary point set;

[0046] The elements in the first boundary point set and the second boundary point set are mapped one by one to form a boundary analysis set: BO = {(bo 1_1 ,bo 2_1 ), (bo 1_2 ,bo 2_1 ),...,(bo 1_n ,bo 2_m )};

[0047] Get the length of the connecting line of each element in the boundary analysis set and define it as the analysis length:

[0048]

[0049] Set a boundary judgment function to determine whether the first sub-region and the second sub-region are adjacent regions:

[0050] If F BO = True, then the two sub-regions are determined to be adjacent; if F BO = False, then the two sub-regions are judged to be non-adjacent;

[0051] Repeat the steps of generating the first sub-region and the second sub-region and performing the judgment until each sub-region in the qualified sub-region set is judged as the first sub-region;

[0052] Obtain the number of the first sub-regions for determining whether two sub-regions are adjacent, recorded as N_Ar;

[0053] Get the overall boundary of the target area, which is defined as the area boundary and recorded as BO_Ar;

[0054] Define a region boundary determination function to determine whether there is a sub-region boundary that coincides with the region boundary within the qualified sub-region set:

[0055]

[0056] Among them, PO Sub-regions Represents the boundary point of the sub-region element in the qualified sub-region set, PO BO_Ar Represents the boundary point of the region boundary, Sub-regions∈Set_Ar represents the sub-region elements in the qualified sub-region set, and Set_Ar represents the qualified sub-region set;

[0057] If F coincide = True, then the boundary of the sub-region is determined to coincide with the region boundary; if F coincide= False, then the boundary of the sub-region is determined to be not coincident with the region boundary;

[0058] Get the number of sub-regions whose boundaries coincide with the region boundary, denoted as N_Co;

[0059] Set a laying requirement judgment function to determine whether the target area can be used for optical cable laying:

[0060] Among them, θγ is the threshold of the sub-region ratio that meets the requirements;

[0061] If F lay = False, it is determined that the target area cannot be used for optical cable laying, and a new target area is selected for laying analysis;

[0062] If F lay = True, the target area is determined to be suitable for optical cable laying. Based on the sub-collection point locations that meet the requirements, several different optical cable laying plans are automatically generated to form an optical cable laying plan set:

[0063] OC_Lay={Lay1, Lay2,..., Lay p}; Each scheme Lay r Contains a collection of waypoints:

[0064] Pol r ={Pol r1 ,Pol r2 ,...,Pol rq}; Among them, Pol rl =(lat rl ,lon rl ,depth rl );

[0065] Set the evaluation index weight vector: W = (w soil , w safety , w resources ); where w soil is the soil condition dimension, w safety is the security dimension, w resources Save dimensions for resources;

[0066] Calculate the overall score for each solution:

[0067] Among them, SoilCondition(Lay r ) is the soil condition score, Safety(Lay r ) is the security score, Resources(Lay r ) to score the resource;

[0068] The solution with the highest score is selected as the final solution:

[0069] As an optional solution to the virtual reality-based intelligent cable laying method of the present invention, the geographical data within the target area is obtained, specifically:

[0070] Get the center point of the target area and set it as the area center point;

[0071] Generate a rectangular area with the center point of the area as the center point so that the rectangular area can completely cover the target area;

[0072] Record the latitude and longitude coordinates of the lower left and upper right corners of the rectangular area to generate the target area range:

[0073] R=(lat min ,lon min ,lat max ,lon max ); where lat min ,lon min is the latitude and longitude of the lower left corner of the target area, lat max ,lon max is the latitude and longitude of the upper right corner of the target area;

[0074] Set up a topographic data interface function to form a set of topographic data points covering the target area R: D = T (lat, lon);

[0075] Among them, each data point contains elevation e, slope s, and soil type t;

[0076] Define a height anomaly threshold range [e min , e max ];

[0077] Traverse the data point set D, for each data point d ij =(lat i ,lon j , e ij , s ij , t ij ), set an elevation value judgment function to determine whether the elevation value of the data point is abnormal:

[0078]

[0079] If F e =1, the elevation value of the data point is judged to be abnormal and marked as an invalid data point: d ij .valid=False;if Fe =0, the elevation value of the data point is judged to be normal;

[0080] Traverse the data point set D, for each data point d ij =(lat i ,lon j , e ij , s ij , t ij ), set a slope value judgment function to determine whether the slope value of the data point is abnormal:

[0081] Among them, s max is the slope anomaly threshold;

[0082] If F s =1, the slope value of the data point is considered abnormal and marked as an invalid data point: d ij .valid=False;if F s =0, the slope value of the data point is judged to be normal;

[0083] Arrange the valid data points after cleaning and verification to form a set of valid data points:

[0084] D clean ={d ij |d ij .valid=True};

[0085] {d ij |d ij .valid=True}={d ij ∈D|e min ≤e ij ≤e max ∧s ij ≤s max}.

[0086] As an optional solution of the optical cable intelligent laying method based on virtual reality of the present invention, the terrain conditions of the target area are judged, specifically: obtaining a set of valid data points D clean ;

[0087] Soil type code set defining soft soil: T soft ={t1, t2, ..., t n};

[0088] Screening soft soil records: D soft ={(lat i ,lon j , e ij , s ij , t ij )∈Dclean |t ij ∈T soft};

[0089] Divide the target area R into m×n grids, and the area of ​​each grid is:

[0090] ΔA = Δlat × Δlon × cos(lat); where Δlat is the latitude interval, Δlon is the longitude interval, and lat is the latitude value;

[0091] Calculate the total area of ​​the target region:

[0092] Calculate the soft soil area:

[0093] Calculate the proportion of soft soil area:

[0094] Set a soft soil judgment function to determine whether the target area is a soft soil zone:

[0095] Among them, is the area ratio threshold of soft soil zone;

[0096] If F ρ = True, the target area is determined to be a soft soil zone; if F ρ = False, it is determined that the target area is not a soft soil zone.

[0097] As an optional solution of the virtual reality-based intelligent cable laying method of the present invention, if the target area is a soft soil zone, a soft soil laying plan is generated, specifically:

[0098] Get the target area range R; get the historical collection record library within the target area range R: H soil ={h s1 , h s2 ,...,h sN}; Each history record h si Contains the geographical coordinates and soil data of the collection points. The geographical coordinates are marked as (lat si ,lon si ), soil data are as follows: d si =(d i1 , d i2 ,...,d im );

[0099] Set a spatial intersection judgment function, for each historical record h si , to determine whether its collection point is within the target area:

[0100] If InRegion = True, the collection point is determined to be within the target area, that is, the historical record and the target area have a spatial intersection; if InRegion = False, the collection point is determined to be not within the target area, that is, the historical record and the target area have no spatial intersection;

[0101] Retrieve all historical collection records that have spatial intersections with the target area to form a historical collection record set: H target ={h si ∈H soil |InRegion(h si )=True};

[0102] Extract historical collection data points and their soil data from the historical collection record set: D hist ={(lat si ,lon si , d si，hist )|h si ∈H target}; where d si，hist Represents historically collected soil data;

[0103] For each historical data point (lat si ,lon si ), set up a soil data collection function to obtain the current soil data: d si，curr =S(lat si ,lon si );

[0104] Create a collection containing historical and current soil data:

[0105] Define a change judgment function to compare the degree of change between historical data and current data:

[0106] Where md is the number of dimensions of soil data;

[0107] For each historical data point (lat si ,lon si ), calculate the degree of change of its soil data:

[0108] Δ si =Change(d si，hist , d si，curr );

[0109] Set a change judgment function to determine whether the change of each data point is large:

[0110]

[0111] Form a set containing the change flags:

[0112]

[0113] Calculate the proportion of data points with large variance: Where N is D change The total number of data points in ;

[0114] Set a collection point judgment function to determine whether the collection point needs to be regenerated:

[0115]

[0116] If Regenerate = True, it is determined that the collection points need to be regenerated;

[0117] If Regenerate=False, it is determined that there is no need to regenerate the collection points, and the normal paving plan is executed.

[0118] As an optional solution of the virtual reality-based intelligent cable laying method of the present invention, if the target area is not a soft soil zone, a common laying plan is generated, specifically, the average elevation of the target area is calculated: Where N is the total number of data points;

[0119] Set the slope s ij Convert to a two-dimensional gradient vector:

[0120] Calculate the average of the gradient vectors of all data points to determine the overall slope trend direction:

[0121]

[0122] Retrieve history plans;

[0123] For candidate solutions, define the weight vector:

[0124] For candidate solutions, calculate the weighted score: Score(h k )=∑ m w m ·v k，m ;

[0125] Set a candidate ranking function and assign all candidate solutions a weighted score Score(h k ) Sort by high to low: H sorted =Sort(H cand ,byScore(h k ), descending);

[0126] Take the solution with the highest score after sorting as the optimal solution: h best =H sorted [0]; Extract the path of the optimal solution, denoted as P best ; The optimal solution path P best Projected to the target area, according to the local slope s ij Make fine adjustments: P adjusted ={AdjustPoint(p, s ij )|p∈P best}; Among them, AdjustPoint(p, s ij ) is the path adjustment function, s ij is the slope of the path point p, where p is the original coordinate of the path point, specifically: p = (lat, lon, depth).

[0127] As an optional solution of the optical cable intelligent laying method based on virtual reality of the present invention, wherein: searching the historical solution is specifically: obtaining the historical experience database: H = [h1, h2, ..., h M ]; where each historical scenario h k Contains the terrain feature vector f k and multi-dimensional evaluation index v k ; Among them, f k Represents the terrain feature vector of the kth historical scheme, specifically: f k =(f k1 , f k2 ,...,f kn ); where v k Represents the multi-dimensional evaluation index vector of the k-th historical plan, specifically: v k =(v k1 , v k2 ,...,v km ); where each historical scheme h k The multi-dimensional evaluation indicators are: k =(v k，cost , v k，convenience , v k，safety );

[0128] Set a soil type set: T = {t T1 , t T2 ,...,t Tn};

[0129] For each soil type t Tk , calculate its clean The proportion of appearance:

[0130] Form the soil type distribution vector:

[0131] Combine the extracted key terrain features into the target area terrain feature vector:

[0132] Define a similarity measurement function to calculate the target area terrain feature vector f target Each historical scenario h in the historical experience database k The terrain feature vector f k Similarities between:

[0133]

[0134] Set a permutation function to sort the historical solutions according to the similarity measure Sim(f target , f k ) to sort: H sorted =Sort(H, bySim(f target , f k ));

[0135] Define a data extraction function to select the top K candidates from the sorted historical solutions to form a candidate set: H cand ={h k1 , h k2 ,...,h kK}; where h k1 , h k2 ,...,h kK It is H sorted The top K historical solutions in .

[0136] As an optional solution to the optical cable intelligent laying method based on virtual reality of the present invention, in which: during the laying process, data is updated and the plan is adjusted in real time, specifically:

[0137] Real-time acquisition of terrain, soil and other data in the construction area: in, It is real-time elevation data. It is real-time slope data. It is real-time soil data;

[0138] For each data point (lat i ,lon j ), compare and analyze the real-time data with the data in the original plan, and calculate the difference between the real-time data and the original plan data: in, These are the elevation, slope, and soil data in the original plan;

[0139] Determine whether the change in each data point is significant:

[0140] Among them, θ e ,θ s and θ t thresholds for changes in elevation, slope, and soil data;

[0141] The proportion of data points with statistically significant changes: Where N is the total number of data points;

[0142] Set an adjustment judgment function to determine whether the solution needs to be updated:

[0143] Among them, θ dat is the significant change ratio threshold;

[0144] If F dat =1, it is determined that the plan needs to be updated, and the updated cable laying plan L updated Contains a new set of waypoints: P updated ={P updated，1 , P updated，2 ,...,P updated，q};

[0145] Among them, P updated，l =(lat updated，l ,lon updated，l ,depth updated，l );

[0146] If F dat =0, it is determined that no update plan is needed.

[0147] The present invention has the following beneficial effects:

[0148] 1. This virtual reality-based intelligent optical cable laying method conducts terrain analysis on the area where the optical cable is to be laid, and determines whether the terrain is soft soil based on historically collected data. If it is not soft soil, a customized optical cable laying plan is generated based on the retrieval and sorting algorithm of the historical experience database and the local optimization adjustment strategy. The topographic and geomorphic data of the target area are obtained using geographic information system technology and related data interfaces, and the accuracy and completeness of the data are ensured by using data cleaning and verification algorithms. Based on the principles of soil type classification standards and area ratio thresholds, the soil type data of the target area is analyzed and judged. Based on detailed topographic and geomorphic data, soil data and historical experience, a more reasonable and feasible optical cable laying plan can be generated. Through precise data acquisition, analysis and comparison, the topographic and geomorphic characteristics and soil conditions of the target area can be accurately judged, thereby improving the pertinence and accuracy of the optical cable laying plan and reducing construction problems and risks caused by unreasonable plans.

[0149] 2. This virtual reality-based intelligent optical cable laying method performs a terrain analysis on the area where the optical cable is to be laid. If the soil is soft, data is collected again from the historical data collection points in the area to determine whether there is a significant change compared with the historical data. If there is a significant change, the target area is divided into several sub-areas, and several data collection points are randomly generated. Then, several sub-collection points of different depths are generated for each data collection point. Data is collected again based on the new collection points. By analyzing the soil data collected at each sub-collection point, it is determined whether there is a sub-area that meets the requirements, and whether the sub-areas that meet the requirements are coherent and consistent. It runs through the target area to determine whether the target area meets the requirements for optical cable laying. It uses soil data collection and analysis, change judgment, re-collection and evaluation technologies to generate and select the optimal optical cable laying plan. It comprehensively considers factors in multiple dimensions such as soil conditions, safety and resource conservation. By optimizing the optical cable laying plan, it can reasonably allocate and utilize construction resources, improve resource utilization efficiency, reduce costs, and achieve the economy and benefits of the optical cable laying project. It can handle optical cable laying problems in ordinary areas as well as special terrain conditions such as soft soil areas. It can meet the optical cable laying needs in different scenarios and has wide applicability and flexibility.

[0150] 3. This intelligent optical cable laying method based on virtual reality determines whether the target area meets the requirements for optical cable laying. If not, the laying area is reselected. During the laying process, data is collected in real time and the plan is adjusted to avoid accidents. With the help of sensors, measuring instruments and other equipment, real-time data is dynamically acquired. The comparative analysis algorithm is used to determine whether the data changes are significant, and the plan adjustment algorithm is used to optimize and update the original laying plan based on the changes. During the optical cable laying process, real-time monitoring and data collection and timely adjustment of the plan can ensure the smooth progress of the construction process, enhance the reliability of optical cable laying, reduce the risk of optical cable damage caused by factors such as changes in the external environment, and extend the service life of the optical cable. BRIEF DESCRIPTION OF THE DRAWINGS

[0151] Figure 1 This is a flow chart of the optical cable intelligent laying method based on virtual reality of the present invention. DETAILED DESCRIPTION

[0152] Example 1: A method for intelligent laying of optical cables based on virtual reality, see Figure 1 ,include:

[0153] Obtain geographic data within the target area;

[0154] If the collection points need to be regenerated, set the target area's regional data;

[0155] At each sub-collection point At , soil data is collected to obtain a soil data set:

[0156] D ijk ={d ijk1 , d ijk2 ,...,d ijkm}; Among them, D ijkt Represents the t-th soil data indicator, such as soil particle composition, water content, carrying capacity, etc.;

[0157] Set the soil data index threshold that meets the requirements: T th ={T th1 , T th2 ,...,T thm}; where T tht represents the threshold value of the t-th soil data indicator;

[0158] For each sub-collection point Define a soil judgment function to determine whether the soil data meets the requirements:

[0159]

[0160] Count each sub-region S i The number of sub-collection points that meet the requirements:

[0161] Set a paving judgment function to judge the sub-area S i Whether it meets the laying requirements:

[0162]

[0163] Among them, α is the proportion threshold that meets the requirements, such as 60%, β is the depth standard deviation threshold, such as 0.2 meters, and S_D_depth is the depth standard deviation;

[0164] If FCT i = True, then the sub-area is determined to meet the paving requirements;

[0165] If FCT i = False, then the sub-area is judged to be not in compliance with the paving requirements;

[0166] Count the number of sub-areas that meet the paving requirements:

[0167] Determine and generate optical cable laying plan.

[0168] Among them, the target area is set for regional data, specifically:

[0169] Get the target area range R;

[0170] Get elevation data e, where e ij Indicates the elevation of the corresponding latitude and longitude point;

[0171] Extract the dimension of elevation data e, recorded as em×en;

[0172] Get the latitude range of the target area [lat min ,lat max ] and longitude range [lon min ,lon max ];

[0173] Calculate the latitude interval between two adjacent elevation data points in the latitude direction:

[0174] Calculate the longitude interval between two adjacent elevation data points in the longitude direction:

[0175] For each grid point (i, j) in the elevation data e, calculate the partial derivatives of the elevation in the latitude and longitude directions:

[0176] Compute terrain gradients based on partial derivatives:

[0177] Combine the latitude and longitude and gradient values ​​of each point into a feature vector: X = [lat, lon, terrain_gradient];

[0178] The feature vectors of all points form a data set Df: Df = {x1, x2, ..., x m};

[0179] Set the number of clusters n, for example, the number of clusters is 5;

[0180] Randomly select the first cluster center c1: c1 = x i , where i is a randomly selected integer between 1 and m;

[0181] For each data point x j , calculate its distance to the nearest cluster center: d j =min 1≤i≤k ||x j -ci||;

[0182] The probability of selecting the next cluster center ck is:

[0183] For each data point x j , calculate its distance to all cluster centers ci||x j -ci||;

[0184] xj Assign to the closest cluster center: cclosest = argmin ci ||x j -ci||;

[0185] For each cluster k, the new cluster center ck new is the mean of all data points in the cluster:

[0186]

[0187] Among them, C k are all data points in cluster k, |C k | is the number of data points in cluster k. A cluster is a collection of data points that are similar to each other in the feature space and very different from data points in other clusters. Clusters reflect the inherent structure and distribution pattern of the data and are natural groupings of data points based on their characteristics. The number, shape, and size of clusters are determined by the data itself and the clustering algorithm. Clusters are the core output of cluster analysis and are used to reveal the structural characteristics of the data.

[0188] For each cluster C k , extract the geographic coordinates of all data points in the cluster:

[0189] C k ={(lat k，1 ,lon k，1 ), (lat k，2 ,lon k，2 ),...,(lat k，nk ,lon k，nk )};

[0190] Where nk is the number of data points in cluster k;

[0191] For each cluster C k , calculate its boundaries, i.e., minimum and maximum latitude and longitude:

[0192]

[0193] Let the boundary of each cluster be represented as a rectangular region:

[0194] S k =(lat min，k ,lon min，k ,lat max，k ,lon max，k );

[0195] Integrate all rectangular areas to form a set of sub-areas: S k ={S1, S2, ..., S n};

[0196] For each sub-region, m collection points are randomly generated inside it, and the collection point set is expressed as:

[0197] P i ={P i1 , P i2 ,...,P im}; Among them, P ij =(lat ij ,lon ij );

[0198] For each collection point P ij , randomly generate k sub-collection points with different depths. The depth range is determined according to the expected depth of the optical cable laying. The sub-collection point set is expressed as:

[0199] in,

[0200] This embodiment also provides for determining and generating an optical cable laying plan, specifically:

[0201] Extract all FCTs i = True, forming a qualified sub-region set;

[0202] Randomly extract any two sub-regions from the qualified sub-region set and define them as the first sub-region and the second sub-region respectively;

[0203] Extracting boundary lines of the first sub-region and the second sub-region respectively, and defining them as the first boundary line and the second boundary line;

[0204] generating a plurality of boundary points on the first boundary line to form a first boundary point set, wherein the boundary points are points on the boundary line, and the boundary line is formed by connecting countless boundary points;

[0205] generating a plurality of boundary points on the second boundary line to form a second boundary point set;

[0206] The elements in the first boundary point set and the second boundary point set are mapped one by one to form a boundary analysis set: BO = {(bo 1_1 ,bo 2_1 ), (bo 1_2 ,bo 2_1 ),...,(bo 1_n ,bo 2_m )};

[0207] Get the length of the connecting line of each element in the boundary analysis set and define it as the analysis length:

[0208]

[0209] Set up a boundary judgment function to determine whether the first sub-area and the second sub-area are adjacent areas. In other words, determine whether the optical cable can be laid continuously as much as possible under the required environment when it is completely laid. In other words, determine whether a relatively coherent laying route can be generated:

[0210]

[0211] If F BO = True, then the two sub-regions are considered adjacent, that is, the first sub-region and the second sub-region have overlapping boundary points;

[0212] If F BO = False, then the two sub-regions are judged to be non-adjacent;

[0213] Repeat the steps of generating the first sub-region and the second sub-region and performing the judgment until each sub-region in the qualified sub-region set is judged as the first sub-region;

[0214] Obtain the number of the first sub-regions for determining whether two sub-regions are adjacent, recorded as N_Ar;

[0215] Get the overall boundary of the target area, which is defined as the area boundary and recorded as BO_Ar;

[0216] Define a region boundary judgment function to determine whether there is a sub-region boundary that coincides with the region boundary within the qualified sub-region set, that is, to determine whether the optical cable can be fully laid through the target area:

[0217]

[0218] Among them, PO Sub-regions Represents the boundary point of the sub-region element in the qualified sub-region set, PO BO_Ar Represents the boundary point of the region boundary, Sub-regions∈Set_Ar represents the sub-region elements in the qualified sub-region set, and Set_Ar represents the qualified sub-region set;

[0219] If F coincide = True, then the boundary of the sub-region is determined to coincide with the region boundary;

[0220] If F coincide = False, then the boundary of the sub-region is determined to be not coincident with the region boundary;

[0221] Get the number of sub-regions whose boundaries coincide with the region boundary, denoted as N_Co;

[0222] Set a laying requirement judgment function to determine whether the target area can be used for optical cable laying:

[0223]

[0224] Among them, θγ is the threshold of the proportion of sub-areas that meet the requirements, such as 70%, which is used to determine whether there are many sub-areas that meet the paving requirements in the target area;

[0225] If F lay = False, it is determined that the target area cannot be used for optical cable laying, and a new target area is selected for laying analysis;

[0226] If F lay = True, the target area is determined to be suitable for optical cable laying, that is, there are many sub-areas that meet the requirements and they are coherent, and the cable can be laid through the target area. Based on the locations of the sub-collection points that meet the requirements, several different optical cable laying plans are automatically generated to form an optical cable laying plan set:

[0227] OC_Lay={Lay1, Lay2,..., Lay p}; Each scheme Lay r Contains a set of path points: Pol r ={Pol r1 ,Pol r2 ,...,Pol rq}; Among them, Pol rl =(lat rl ,lon rl ,depth rl );

[0228] Set the evaluation index weight vector: W = (w soil , w safety , w resources );

[0229] Among them, w soil is the soil condition dimension, w safety is the security dimension, w resources It is a resource-saving dimension used to evaluate the schemes in combination with multiple dimensions and select the best scheme as the final paving scheme;

[0230] Calculate the overall score for each solution:

[0231] Score(Lay r )=w soil Soil Condition (Lay r )+

[0232] w safety Safety(Lay r )+w resources Resources(Lay r);

[0233] Among them, SoilCondition(Lay r ) is a soil condition score used to evaluate the cable laying plan based on factors such as soil particle composition, moisture content, bearing capacity and pH. r Is the soil characteristics of the selected route suitable for laying optical cables? r ) is a safety score used to evaluate the cable laying plan based on factors such as the steepness of the terrain, landslide risk, and seismic activity. r Safety assessment of selected path points, Resources(Lay r ) is a resource score, which is used to evaluate the cable laying plan based on factors such as material consumption, manpower input, and mechanical equipment usage required for construction. r Assessment of resource efficiency required;

[0234] The solution with the highest score is selected as the final solution:

[0235] Example 2: This example is an improvement made on the basis of Example 1. The optical cable intelligent laying method based on virtual reality obtains geographic data within the target area, specifically:

[0236] Get the center point of the target area and set it as the area center point;

[0237] Generate a rectangular area with the center point of the area as the center point so that the rectangular area can completely cover the target area;

[0238] Record the latitude and longitude coordinates of the lower left and upper right corners of the rectangular area to generate the target area range:

[0239] R=(lat min ,lon min ,lat max ,lon max ); where lat min ,lon min is the latitude and longitude of the lower left corner of the target area, lat max ,lon max is the latitude and longitude of the upper right corner of the target area;

[0240] Set up a topographic data interface function to form a set of topographic data points covering the target area R: D = T (lat, lon);

[0241]

[0242] Among them, each data point contains key information such as elevation e, slope s, soil type t, etc.

[0243] Define a height anomaly threshold range [e min , e max ];

[0244] Traverse the data point set D, for each data point d ij =(lat i ,lon j , e ij , s ij , t ij ), set an elevation value judgment function to determine whether the elevation value of the data point is abnormal:

[0245]

[0246] If F e =1, the elevation value of the data point is considered abnormal and marked as an invalid data point:

[0247] d ij .valid=False;

[0248] If F e =0, the elevation value of the data point is judged to be normal;

[0249] Traverse the data point set D, for each data point d ij =(lat i ,lon j , e ij , s ij , t ij ), set a slope value judgment function to determine whether the slope value of the data point is abnormal:

[0250] Among them, s max is the slope anomaly threshold, which is used to determine whether the slope value of a data point is abnormal;

[0251] If F s =1, the slope value of the data point is considered abnormal and marked as an invalid data point:

[0252] d ij .valid=False;

[0253] If F s =0, the slope value of the data point is judged to be normal;

[0254] Arrange the valid data points after cleaning and verification to form a set of valid data points:

[0255] D clean={d ij |d ij .valid=True};

[0256] {d ij |d ij .valid=True}={d ij ∈D|e min ≤e ij ≤e max ∧s ij ≤s max}.

[0257] This embodiment also provides for determining the terrain conditions of the target area, specifically:

[0258] Get the valid data point set D clean ;

[0259] Define the soil type code set for soft soil. For example, if the soil type code is 5, it represents soft soil. Then T soft ={5}:T soft ={t1, t2, ..., t n};

[0260] Screening soft soil records: D soft ={(lat i ,lon j , e ij , s ij , t ij )∈D clean |t ij ∈T soft};

[0261] Divide the target area R into m×n grids, and the area of ​​each grid is:

[0262] ΔA = Δlat × Δlon × cos(lat); where Δlat is the latitude interval, Δlon is the longitude interval, and lat is the latitude value;

[0263] Calculate the total area of ​​the target region:

[0264] Calculate the soft soil area:

[0265] Calculate the proportion of soft soil area:

[0266] Set a soft soil judgment function to determine whether the target area is a soft soil zone:

[0267] Among them, is the area ratio threshold of the soft soil zone, which is used to determine whether the target area is a soft soil zone. For example, the area ratio threshold of the soft soil zone is 30%;

[0268] If F ρ = True, the target area is determined to be a soft soil zone;

[0269] If F ρ = False, it is determined that the target area is not a soft soil zone.

[0270] If the target area is a soft soil zone, a soft soil paving plan is generated, specifically:

[0271] Get the historical collection record library within the target area R: H soil ={h s1 , h s2 ,...,h sN}; Each history record h si Contains the geographical coordinates of the collection points and soil data, such as soil particle composition, water content, carrying capacity, etc. The geographical coordinates are marked as (lat si ,lon si ), soil data are as follows:

[0272] d si =(d i1 , d i2 ,...,d im );

[0273] Set a spatial intersection judgment function, for each historical record h si , to determine whether its collection point is within the target area:

[0274]

[0275] If InRegion = True, the collection point is determined to be within the target area, that is, the historical record has a spatial intersection with the target area;

[0276] If InRegion = False, it is determined that the collection point is not in the target area, that is, the historical record has no spatial intersection with the target area;

[0277] Retrieve all historical collection records that have spatial intersections with the target area to form a historical collection record set: H target ={h si ∈H soil |InRegion(h si )=True};

[0278] Extract historical collection data points and their soil data from the historical collection record set:

[0279] D hist ={(lat si ,lon si , d si，hist )|h si ∈H target}; where d si，hist Represents historically collected soil data;

[0280] For each historical data point (lat si ,lon si ), set up a soil data collection function to obtain the current soil data: d si，curr =S(lat si ,lon si );

[0281] Create a collection containing historical and current soil data:

[0282]

[0283] Define a change judgment function to compare the degree of change between historical data and current data:

[0284] Where md is the dimensionality of the soil data, that is, the number of different attributes or features contained in the soil data, such as soil particle composition, water content, carrying capacity, etc., which are used to analyze the soil condition;

[0285] For each historical data point (lat si ,lon si ), calculate the degree of change of its soil data:

[0286] Δ si =Change(d si，hist , d si，curr );

[0287] Set a change judgment function to determine whether the change of each data point is large:

[0288]

[0289] Form a set containing the change flags:

[0290]

[0291] Calculate the proportion of data points with large variance: Where N is D change The total number of data points in ;

[0292] Set a collection point judgment function to determine whether the collection point needs to be regenerated:

[0293]

[0294] If Regenerate = True, it is determined that the collection points need to be regenerated;

[0295] If Regenerate=False, it is determined that there is no need to regenerate the collection points, and the normal paving plan is executed.

[0296] This embodiment also provides that if the target area is not a soft soil zone, a normal paving plan is generated, specifically:

[0297] Get the valid data point set D clean ;

[0298] Calculate the average elevation of the target area: Where N is the total number of data points;

[0299] Set the slope s ij Convert to a two-dimensional gradient vector:

[0300] Calculate the average of the gradient vectors of all data points to determine the overall slope trend direction:

[0301]

[0302] Retrieve history plans;

[0303] For candidate solutions, define the weight vector: Among them, the weight value range is between [0, 1] and satisfies w cost +w convenience +w safety =1;

[0304] For candidate solutions, calculate the weighted score: Score(h k )=∑ m w m ·v k，m ;

[0305] Set a candidate ranking function and assign all candidate solutions a weighted score Score(h k ) Sort by high to low: H sorted =Sort(H cand ,byScore(h k ), descending);

[0306] Take the solution with the highest score after sorting as the optimal solution: h best =H sorted [0];

[0307] Extract the path of the optimal solution, denoted as Pbest ;

[0308] The optimal solution path P best Projected to the target area, according to the local slope s ij Make fine adjustments:

[0309] P adjusted ={AdjustPoint(p, s ij )|p∈P best}; Among them, AdjustPoint(p, s ij ) is the path adjustment function, which is used to adjust the local slope s of the target area ij Adjust the location of the path points to reduce the difficulty of construction, s ij is the slope of the path point p, where p is the original coordinate of the path point, including latitude, longitude and laying depth, specifically:

[0310] p = (lat, lon, depth).

[0311] Among them, the retrieval history plan is as follows:

[0312] Obtain historical experience database: H = [h1, h2, ..., h M ]; where each historical scenario h k Contains the terrain feature vector f k and multi-dimensional evaluation index v k ; Among them, f k Represents the terrain feature vector of the kth historical scheme, specifically: f k =(f k1 , f k2 ,...,f kn );

[0313] Among them, v k Represents the multi-dimensional evaluation index vector of the k-th historical solution, specifically:

[0314] v k =(v k1 , v k2 ,...,v km );

[0315] Among them, each historical plan h k The multi-dimensional evaluation indicators are as follows:

[0316] v k =(v k，cost , v k，convenience , v k，safety ); where v k，cost 、v k，convenience and v k，safetyThey represent the evaluation indicators of the three dimensions of cost, construction convenience and safety assurance;

[0317] Set a soil type set: T = {t T1 , t T2 ,...,t Tn};

[0318] For each soil type t Tk , calculate its clean The proportion of appearance:

[0319] Form the soil type distribution vector:

[0320] Combine the extracted key terrain features into the target area terrain feature vector:

[0321]

[0322] Define a similarity measurement function to calculate the target area terrain feature vector f target Each historical scenario h in the historical experience database k The terrain feature vector f k Similarities between:

[0323]

[0324] Set a permutation function to sort the historical solutions according to the similarity measure Sim(f target , f k ) to sort: H sorted =Sort(H, bySim(f target , f k ));

[0325] Define a data extraction function to select the top K candidates from the sorted historical solutions to form a candidate set: H cand ={h k1 , h k2 ,...,h kK}; where h k1 , h k2 ,...,h kK It is H sorted The top K historical solutions in .

[0326] This embodiment also provides real-time data updating and plan adjustment during the laying process, specifically:

[0327] Real-time acquisition of terrain, soil and other data in the construction area:

[0328] in, It is real-time elevation data. It is real-time slope data. It is real-time soil data;

[0329] For each data point (lat i ,lon j ), compare and analyze the real-time data with the data in the original plan, and calculate the difference between the real-time data and the original plan data:

[0330]

[0331] in, These are the elevation, slope, and soil data in the original plan;

[0332] Determine whether the change in each data point is significant:

[0333]

[0334] Among them, θ e ,θ s and θ t It is the change threshold of elevation, slope and soil data, used to judge whether the data change is large;

[0335] The proportion of data points with statistically significant changes: Where N is the total number of data points;

[0336] Set an adjustment judgment function to determine whether the solution needs to be updated:

[0337] Among them, θ dat The significant change ratio threshold, such as 10%, is used to determine whether the data change is large;

[0338] If F dat =1, it is determined that the plan needs to be updated. For data points with significant changes, if the soil bearing capacity decreases, the cable laying depth is increased or reinforcement measures are added. If the terrain changes, the cable path is adjusted to avoid the affected area. If the slope changes, the cable path is adjusted to reduce the construction difficulty. The updated cable laying plan L updated Contains a new set of waypoints:

[0339] P updated ={P updated，1 , P updated，2 ,...,P updated，q};

[0340] Among them, P updated，l =(lat updated，l ,lon updated，l ,depthupdated，l );

[0341] If F dat =0, it is determined that no update plan is needed.

[0342] This embodiment reduces construction problems and risks caused by unreasonable plans, ensures the smooth progress of the construction process, enhances the reliability of optical cable laying, reduces the risk of optical cable damage caused by factors such as changes in the external environment, and extends the service life of the optical cable. It comprehensively considers factors in multiple dimensions such as soil conditions, safety, and resource conservation. By optimizing the optical cable laying plan, it can reasonably allocate and utilize construction resources, improve resource utilization efficiency, reduce costs, and achieve the economy and benefits of the optical cable laying project. It can handle optical cable laying problems in ordinary areas as well as special terrain conditions such as soft soil areas. It can meet the optical cable laying needs in different scenarios and has wide applicability and flexibility.

Claims

1. A method for intelligent laying of optical cables based on virtual reality, characterized in that: include: Obtain geographic data within the target area; If the collection points need to be regenerated, set the target area's regional data; At each sub-collection point At , soil data is collected to obtain a soil data set: D ijk ={d ijk1 ,d ijk2 ,...,d ijkm }; Among them, D ijkt represents the t-th soil data indicator; Set the soil data indicator thresholds that meet the requirements: T th ={T th1 ,T th2 ,...,T thm }; Among them, T tht represents the threshold value of the t-th soil data indicator; For each sub-collection point Define a soil judgment function to determine whether the soil data meets the requirements: Count each sub-region S i The number of sub-collection points that meet the requirements: Set a paving judgment function to judge the sub-area S i Whether it meets the laying requirements: Among them, α is the required ratio threshold, β is the depth standard deviation threshold, and S_D_depth is the depth standard deviation; If FCT i = True, then the sub-area is determined to meet the paving requirements; If FCT i = False, then the sub-area is judged to be not in compliance with the paving requirements; Count the number of sub-areas that meet the paving requirements: Determine and generate optical cable laying plan.

2. The optical cable intelligent laying method based on virtual reality according to claim 1, characterized in that: Set the target area's regional data as follows: Get the target area range R; Get elevation data e; Extract the dimension of elevation data e, recorded as em×en; Get the latitude range of the target area [lat min ,lat max ] and longitude range [lon min ,lon max ]; Calculate the latitude interval between two adjacent elevation data points in the latitude direction: Calculate the longitude interval between two adjacent elevation data points in the longitude direction: For each grid point (i, j) in the elevation data e, calculate the partial derivatives of the elevation in the latitude and longitude directions: Compute terrain gradients based on partial derivatives: Combine the latitude and longitude and gradient value of each point into a feature vector: X=[lat,lon,terrain_gradient]; The feature vectors of all points form a data set Df: Df={x1,x2,...,x m }; Set the number of clusters n; Randomly select the first cluster center c1: c1 = x i ; For each data point x j , calculate its distance to the nearest cluster center: d j =min 1≤i≤k ||x j m||; The probability of selecting the next cluster center ck is: For each data point x j , calculate its distance to all cluster centers ci||x j -ci||; x j Assign to the closest cluster center: cclosest=argmin ci ||x j -ci||; For each cluster k, the new cluster center ck new is the mean of all data points in the cluster: Among them, C k are all data points in cluster k, |C k | is the number of data points in cluster k; For each cluster C k , extract the geographic coordinates of all data points in the cluster: C k ={(years k，1 ,lon k，1 ),(years k，2 ,lon k，2 ),...,(years k，nk ,lon k，nk )}; Where nk is the number of data points in cluster k; For each cluster C k , calculate its bounds: Let the boundary of each cluster be represented as a rectangular region: S k =(years min，k ,lon min，k ,years max，k ,lon max，k ); Integrate all rectangular regions to form a set of subregions: S k ={S1,S2,...,S n }; For each sub-region, m collection points are randomly generated inside it, and the collection point set is expressed as: P i ={P i1 ,P i2 ,...,P im }; Among them, P ij =(lat ij ,lon ij ); For each collection point P ij , randomly generate k sub-collection points with different depths, then the sub-collection point set is expressed as: in, 3. The optical cable intelligent laying method based on virtual reality according to claim 2, characterized in that: Determine and generate the optical cable laying plan, specifically: Extract all FCTs i = True, forming a qualified sub-region set; Randomly extract any two sub-regions from the qualified sub-region set and define them as the first sub-region and the second sub-region respectively; Extracting boundary lines of the first sub-region and the second sub-region respectively, and defining them as the first boundary line and the second boundary line; generating a plurality of boundary points on the first boundary line to form a first boundary point set; generating a plurality of boundary points on the second boundary line to form a second boundary point set; Map the elements in the first boundary point set to the elements in the second boundary point set one by one to form a boundary analysis set: BO={(bo 1_1 , is 2_1 ),(bo 1_2 , is 2_1 ),...,(be 1_n , is 2_m )}; Get the length of the connecting line of each element in the boundary analysis set and define it as the analysis length: Set a boundary judgment function to determine whether the first sub-region and the second sub-region are adjacent regions: If F BO = True, then the two sub-regions are judged to be adjacent; If F BO = False, then the two sub-regions are judged to be not adjacent; Repeat the steps of generating the first sub-region and the second sub-region and performing the judgment until each sub-region in the qualified sub-region set is judged as the first sub-region; Obtain the number of the first sub-regions for determining whether two sub-regions are adjacent, recorded as N_Ar; Get the overall boundary of the target area, which is defined as the area boundary and recorded as BO_Ar; Define a region boundary determination function to determine whether there is a sub-region boundary that coincides with the region boundary within the qualified sub-region set: Among them, PO Sub-regions Represents the boundary point of the sub-region element in the qualified sub-region set, PO BO_Ar Represents the boundary point of the region boundary, Sub-regions∈Set_Ar represents the sub-region elements in the qualified sub-region set, and Set_Ar represents the qualified sub-region set; If F coincide = True, then the boundary of the sub-region is determined to coincide with the region boundary; If F coincide = False, then the boundary of the sub-region is determined to be not coincident with the region boundary; Get the number of sub-regions whose boundaries coincide with the region boundary, denoted as N_Co; Set a laying requirement judgment function to determine whether the target area can be used for optical cable laying: Among them, θγ is the threshold of the sub-region ratio that meets the requirements; If F lay = False, it is determined that the target area cannot be used for optical cable laying, and a new target area is selected for laying analysis; If F lay = True, the target area is determined to be suitable for optical cable laying. Based on the sub-collection point locations that meet the requirements, several different optical cable laying plans are automatically generated to form an optical cable laying plan set: OC_Lay={Lay1,Lay2,...,Lay p }; Among them, each scheme Lay r Contains a collection of waypoints: Pol r ={Pol r1 ,Pol r2 ,...,Pol rq }; Among them, Pol rl =(lat rl ,lon rl ,depth rl ); Set the evaluation indicator weight vector: In=(in soil ,In safety ,In resources ); Among them, w soil is the soil condition dimension, w safety is the security dimension, w resources Save dimensions for resources; Calculate the overall score for each solution: Among them, SoilCondition(Lay r ) is the soil condition score, Safety(Lay r ) is the security score, Resources(Lay r ) to score the resource; The solution with the highest score is selected as the final solution:

4. The optical cable intelligent laying method based on virtual reality according to claim 1, characterized in that: Get geographic data within the target area, specifically: Get the center point of the target area and set it as the area center point; Generate a rectangular area with the center point of the area as the center point so that the rectangular area can completely cover the target area; Record the latitude and longitude coordinates of the lower left and upper right corners of the rectangular area to generate the target area range: R=(years min ,lon min ,years max ,lon max ); Among them, lat min ,lon min is the latitude and longitude of the lower left corner of the target area, lat max ,lon max is the latitude and longitude of the upper right corner of the target area; Set up a terrain data interface function to form a set of terrain data points covering the target area R: D = T (lat, lon); Among them, each data point contains elevation e, slope s, and soil type t; Define a height anomaly threshold range [e min , e max ]; Traverse the data point set D, for each data point d ij =(lat i ,lon j , e ij , s ij , t ij ), set an elevation value judgment function to determine whether the elevation value of the data point is abnormal: If F e =1, the elevation value of the data point is considered abnormal and marked as an invalid data point: d ij .valid=False; If F e =0, the elevation value of the data point is judged to be normal; Traverse the data point set D, for each data point d ij =(lat i ,lon j , e ij , s ij , t ij ), set a slope value judgment function to determine whether the slope value of the data point is abnormal: Among them, s max is the slope anomaly threshold; If F s =1, the slope value of the data point is considered abnormal and marked as an invalid data point: d ij .valid=False; If F s =0, the slope value of the data point is judged to be normal; Arrange the valid data points after cleaning and verification to form a set of valid data points: D clean ={d ij |d ij .valid=True}; {d ij |d ij .valid=True}={d ij ∈D|e min ≤e ij ≤e max ∧s ij ≤s max }。 5. The optical cable intelligent laying method based on virtual reality according to claim 4, characterized in that: Determine the terrain conditions of the target area, specifically: Get the valid data point set D clean ; A set of soil type codes defining soft soils: T soft ={t1,t2,...,t n }; Filter soft soil records: D soft ={(lat i ,lon j ,e ij ,s ij ,t ij )∈D clean |t ij ∈T soft }; Divide the target area R into m×n grids, and the area of ​​each grid is: ΔA=Δlat×Δlon×cos(lat); Among them, Δlat is the latitude interval, Δlon is the longitude interval, and lat is the latitude value; Calculate the total area of ​​the target region: Calculate the soft soil area: Calculate the proportion of soft soil area: Set a soft soil judgment function to determine whether the target area is a soft soil zone: Among them, is the area ratio threshold of soft soil zone; If F ρ = True, the target area is determined to be a soft soil zone; If F ρ = False, it is determined that the target area is not a soft soil zone.

6. The optical cable intelligent laying method based on virtual reality according to claim 5, characterized in that: If the target area is a soft soil zone, a soft soil paving plan is generated, specifically: Get the target area range R; Get the historical collection record library within the target area range R: H soil ={h s1 ,h s2 ,...,h sN }; Among them, each history h si Contains the geographical coordinates and soil data of the collection points. The geographical coordinates are marked as (lat si ,lon si ), soil data are as follows: d si =(d i1 ,d i2 ,...,d im ); Set a spatial intersection judgment function, for each historical record h si , to determine whether its collection point is within the target area: If InRegion = True, the collection point is determined to be within the target area, that is, the historical record has a spatial intersection with the target area; If InRegion = False, it is determined that the collection point is not in the target area, that is, the historical record has no spatial intersection with the target area; Retrieve all historical collection records that have spatial intersections with the target area to form a historical collection record set: H target ={h si ∈H soil |InRegion(h si )=True}; Extract historical collection data points and their soil data from the historical collection record set: D hist ={(lat si ,lon si ,d si，hist )|h si ∈H target }; Among them, d si，hist Represents historically collected soil data; For each historical data point (lat si ,lon si ), set up a soil data collection function to obtain the current soil data: d si，curr =S(lat si ,lon si ); Create a collection containing historical and current soil data: Define a change judgment function to compare the degree of change between historical data and current data: Where md is the number of dimensions of soil data; For each historical data point (lat si ,lon si ), calculate the degree of change of its soil data: Δ si =Change(d si，hist , d si，curr ); Set a change judgment function to determine whether the change of each data point is large: Form a set containing the change flags: Calculate the proportion of data points with large variance: Where N is D change The total number of data points in ; Set a collection point judgment function to determine whether the collection point needs to be regenerated: If Regenerate = True, it is determined that the collection points need to be regenerated; If Regenerate=False, it is determined that there is no need to regenerate the collection points, and the normal paving plan is executed.

7. The optical cable intelligent laying method based on virtual reality according to claim 6, characterized in that: If the target area is not a soft soil zone, a normal paving plan is generated, specifically: Get the valid data point set D clean ; Calculate the average elevation of the target area: Where N is the total number of data points; Set the slope s ij Convert to a two-dimensional gradient vector: Calculate the average of the gradient vectors of all data points to determine the overall slope trend direction: Retrieve history plans; For candidate solutions, define the weight vector: For candidate solutions, calculate the weighted score: Score(h k )=∑ m w m ·v k，m ; Set a candidate ranking function and assign all candidate solutions a weighted score Score(h k ) Sort by high to low: H sorted =Sort(H cand ,by Score(h k ),descending); Take the solution with the highest score after sorting as the optimal solution: h best =H sorted [0]; Extract the path of the optimal solution, denoted as P best ; The optimal solution path P best Projected to the target area, according to the local slope s ij Make fine adjustments: P adjusted ={AdjustPoint(p,s ij )|p∈P best }; Among them, AdjustPoint(p,s ij ) is the path adjustment function, s ij is the slope of the path point p, and p is the original coordinate of the path point, specifically: p = (lat, lon, depth).

8. The optical cable intelligent laying method based on virtual reality according to claim 7, characterized in that: Search history plan, specifically: Get the historical experience database: H=[h1,h2,...,h M ]; Among them, each historical plan h k Contains the terrain feature vector f k and multi-dimensional evaluation index v k ; Among them, f k Represents the terrain feature vector of the k-th historical scheme, specifically: f k =(f k1 ,f k2 ,...,f kn ); Among them, v k Represents the multi-dimensional evaluation index vector of the k-th historical solution, specifically: v k =(v k1 ,v k2 ,...,v km ); Among them, each historical plan h k The multi-dimensional evaluation indicators are as follows: v k =(v k，cost ,v k，convenience ,v k，safety ); Set up a soil type collection: T={t T1 ,t T2 ,...,t Tn }; For each soil type t Tk , calculate its clean The proportion of appearance: Form the soil type distribution vector: Combine the extracted key terrain features into the target area terrain feature vector: Define a similarity measurement function to calculate the target area terrain feature vector f target Each historical scenario h in the historical experience database k The terrain feature vector f k Similarities between: Set a permutation function to sort the historical solutions according to the similarity measure Sim(f target , f k ) to sort: H sorted =Sort(H,bySim(f target ,f k )); Define a data extraction function to select the top K candidates from the sorted historical solutions to form a candidate set: H cand ={h k1 ,h k2 ,...,h kK }; Among them, h k1 , h k2 ,...,h kK It is H sorted The top K historical solutions in .

9. The optical cable intelligent laying method based on virtual reality according to claim 1, characterized in that: During the laying process, data is updated in real time and the plan is adjusted, specifically: Real-time acquisition of terrain, soil and other data in the construction area: in, It is real-time elevation data. It is real-time slope data. It is real-time soil data; Analyze and calculate the difference between real-time data and original solution data: in, These are the elevation, slope, and soil data in the original plan; Determine whether the change in each data point is significant: Among them, θ e ,θ s and θ t thresholds for changes in elevation, slope, and soil data; The proportion of data points with statistically significant changes: Where N is the total number of data points; Set an adjustment judgment function to determine whether the solution needs to be updated: Among them, θ dat is the significant change ratio threshold; If F dat =1, it is determined that the plan needs to be updated, and the updated cable laying plan L updated Contains a new set of waypoints: If F dat =0, it is determined that no update plan is needed.

Images