Surveying and mapping task path planning method and system based on big data

Abstract

The invention belongs to the technical field of surveying and mapping task path planning, and particularly discloses a surveying and mapping task path planning method and system based on big data, and the method comprises the steps: firstly obtaining the current and historical surveying and mapping data of a target region, and recognizing an abnormal change pattern spot through calculating the change confidence; judging whether the pattern spots have geographic information changes or not based on a preset rule, generating variable surveying and mapping priorities, performing surveying and mapping zoning by utilizing a spatial clustering algorithm, generating a preliminary surveying and mapping task path in combination with a sequence optimization algorithm, and importing the path into a geographic environment model to perform safety conflict detection and multi-target efficiency analysis; and finally, correcting the path according to a detection analysis result and outputting a final surveying and mapping task path. Through quick discovery and efficient response to geographic information change, the generated surveying and mapping task path has high safety and high efficiency, and the input-output ratio and the operation intelligence level of surveying and mapping resources are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of surveying and mapping task path planning technology, and relates to a surveying and mapping task path planning method and system based on big data. Background Technology

[0002] With the rapid development of remote sensing technology, UAV mapping, and the spatial information industry, how to efficiently and accurately acquire and update geographic information data has become a core requirement in related fields. Traditional mapping task path planning methods usually rely on preset fixed routes or regional division based on human experience, lacking the ability to perceive and respond to dynamic changes in mapping targets. This makes it difficult to accurately allocate limited mapping resources to the areas that most urgently need updating when performing large-scale, periodic geographic information update mapping.

[0003] For example, Chinese invention patent CN111986193B discloses a method, electronic device, and storage medium for detecting changes in remote sensing images. This method processes input remote sensing images by constructing a deep learning model that includes two modules: depth feature extraction and depth feature fusion classification of irregular image objects, and finally outputs a change detection result map. This method uses an unsupervised stacked denoising autoencoder for pre-training, which reduces the model's dependence on a large amount of labeled data and can preserve the edge and shape information of irregular objects.

[0004] For example, Chinese invention patent CN118395367A discloses a multi-fusion mapping method and system based on UAV aerial surveying and GPS-RTK. This scheme collects environmental data around the target area to construct a feature model, and then generates an optimized flight path for the UAV based on this model. During the planning process, it comprehensively considers various environmental factors such as geographical location, weather conditions, airspace restrictions, and terrain, and continuously calibrates the mapping equipment through a real-time calibration mechanism, aiming to improve the safety of the path and the accuracy of the mapping data.

[0005] The existing technologies mentioned above have the following shortcomings: 1. Currently, they mainly focus on the change detection of remote sensing images themselves, and their output results stop at the change area map. They fail to effectively link with the subsequent surveying and mapping task execution stage and lack a mechanism to transform the change detection results into priority surveying and mapping tasks and further drive path planning. As a result, the changes they identify cannot automatically and efficiently guide the accurate deployment of field surveying and mapping resources.

[0006] 2. Currently, safe flight paths are generated by collecting surrounding environmental data. However, the path planning is based on a preset, static surveying area. It fails to introduce dynamically identified geographic information change areas as the driving source for path planning. At the same time, it fails to assign differentiated priorities to surveying tasks according to the urgency and importance of the changes. As a result, the planned paths lack pertinence and responsiveness when dealing with the dynamic update needs of geographic information. Summary of the Invention

[0007] In view of this, in order to solve the problems mentioned in the background technology, a method and system for surveying and mapping task path planning based on big data is proposed.

[0008] The objective of this invention can be achieved through the following technical solution: This invention provides a mapping task path planning method based on big data, including: S1, acquiring the current mapping data of the target area, comparing it with historical mapping data to calculate the change confidence, and identifying abnormal change patches accordingly.

[0009] S2. Based on the preset change judgment rules, determine whether there are changes in geographic information in the abnormal change patches. If so, assess the urgency of the change based on the change confidence level and generate the change mapping priority for each abnormal change patch.

[0010] S3. Based on the change mapping priority, the abnormal change patches are grouped into mapping partitions by spatial clustering algorithm, and a preliminary mapping task path is generated by sequence optimization algorithm accordingly.

[0011] S4. Import the preliminary mapping task path into the geographic environment model and perform path safety conflict detection and multi-objective efficiency analysis.

[0012] S5. Based on the detection and analysis results, determine whether the preliminary surveying task path needs to be corrected. If so, correct the preliminary surveying task path to obtain the final surveying task path; otherwise, use the preliminary surveying task path as the final surveying task path.

[0013] The present invention also provides a big data-based surveying and mapping task path planning system, including: a change patch identification module, which acquires the current surveying and mapping data of the target area, compares it with historical surveying and mapping data to calculate the change confidence level, and identifies abnormal change patches accordingly.

[0014] The change priority generation module determines whether there are geographic information changes in abnormal change patches based on preset change judgment rules. If so, it assesses the urgency of the change based on the change confidence level and generates the change mapping priority for each abnormal change patch.

[0015] The initial mapping path module, based on the changed mapping priority, uses a spatial clustering algorithm to group abnormally changed patches to form mapping partitions, and then uses a sequence optimization algorithm to generate a preliminary mapping task path.

[0016] The path conflict detection module imports the preliminary mapping task path into the geographic environment model to perform path safety conflict detection and multi-objective efficiency analysis.

[0017] The task path decision module determines whether the preliminary surveying task path needs to be corrected based on the detection and analysis results. If so, the preliminary surveying task path is corrected to obtain the final surveying task path; otherwise, the preliminary surveying task path is used as the final surveying task path.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By deeply integrating change detection, priority evaluation and path planning, the present invention enables the mapping path to directly respond to dynamically identified changes in geographic information, overcomes the drawback of traditional static planning being out of touch with the actual situation, and significantly improves the timeliness of geographic information updates and the ability to actively respond to external changes.

[0019] (2) This invention generates a quantitative change confidence score by normalizing and weighting the comprehensive area change rate, land cover type conversion degree and texture feature variation degree. This index not only improves the reliability of the determination of the authenticity of the change, but also provides a precise numerical decision basis for assessing the urgency of the change and the priority of the task.

[0020] (3) This invention sorts abnormally changed patches by introducing the change mapping impact degree, and forms a two-level optimization sequence of partitions and patches based on spatial clustering, which ensures that mapping resources can be prioritized to the areas with the most significant changes and the greatest impact, thereby maximizing the benefits of mapping operations under the condition of limited resources.

[0021] (4) By introducing a dual verification process of security conflict detection and multi-objective efficiency analysis after path generation, this invention can perform local replanning for security conflicts or sequence re-optimization for efficiency issues. Through iterative correction, it ensures that the final mapping task path has both high security and high efficiency in complex environments. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments 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.

[0023] Figure 1This is a schematic diagram showing the connections between the steps of the method of the present invention.

[0024] Figure 2 This is a schematic diagram showing the connection steps of the abnormal change patch identification method in this invention.

[0025] Figure 3 This is a schematic diagram showing the connections of the various modules in the system of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1 As shown, the present invention provides a mapping task path planning method based on big data. The method includes: S1, obtaining the current mapping data of the target area, comparing it with historical mapping data to calculate the change confidence, and identifying abnormal change patches accordingly.

[0028] Please see Figure 2 As shown, for example, the identification of abnormal change patches includes: extracting the near-infrared and red light band reflectance values ​​of each pixel from the current mapping data and the historical mapping data respectively, calculating the normalized vegetation index based on the reflectance values, and calculating the difference between the two normalized vegetation indices as the degree of difference.

[0029] It should be added that the normalized vegetation index is calculated as follows: using the formula... Calculation of Normalized Difference Vegetation Index ,in, Represents the reflectance value of a pixel in the near-infrared band. This represents the reflectivity value of a pixel in the red light band.

[0030] The value ranges from -1 to +1. Positive values ​​generally indicate vegetation cover, with larger values ​​representing more lush vegetation. Values ​​close to zero indicate rock or bare soil, while negative values ​​may indicate clouds, water, or snow.

[0031] Prioritizing reflectance values ​​in the near-infrared and red light bands is based on the physical mechanism that vegetation and non-vegetated features exhibit significant differences in their spectral responses in these two bands. Healthy vegetation has high reflectance in the near-infrared band, while exhibiting low reflectance in the red light band due to chlorophyll absorption. Conversely, bare soil, water bodies, and buildings show distinctly different reflectance characteristics in these two bands. Therefore, the normalized vegetation index calculated using these two bands is a highly sensitive indicator for identifying land cover status and changes.

[0032] Calculate the average of the absolute values ​​of the differences of all pixels, and identify the set of pixels whose absolute values ​​of differences are greater than the average and form a continuous area in space as candidate change areas.

[0033] Obtain the current area and historical area of ​​the candidate change region, and calculate their relative change rate as the area change rate.

[0034] The area proportion of all dominant land cover types in the candidate change area is extracted from the current and historical survey data, the change rate of the area proportion of each dominant land cover type in the two time phases is calculated, and the change rate of the area proportion is weighted and fused to obtain the land cover type conversion degree.

[0035] The dominant land cover types mentioned in this invention refer to several land cover categories that, based on the specific surveying and mapping task requirements, have the highest area proportion or significant monitoring importance within the candidate change area. For example, the dominant land cover types include, but are not limited to: vegetation types (e.g., forest land, grassland, cultivated land); man-made surface types (e.g., urban residential land, industrial land, transportation land); water areas (e.g., rivers, lakes, reservoirs, ponds); and natural surface types (e.g., bare land, sandy land, saline-alkali land).

[0036] It should be added that the weighted fusion calculation formula for the land cover type conversion degree is as follows: .

[0037] In the formula For land cover type conversion degree, For the first The absolute value of the rate of change in the area proportion of the dominant land cover type. For the first The weight of the rate of change in the area share of dominant land cover types. The total number of dominant land cover types.

[0038] By using weighted fusion to calculate the degree of land cover type transformation, on the one hand, the weight allocation can reflect the different degrees of impact of different land cover type transformations on changes in geographic information, and reflect the differences in the contribution of various land cover changes to changes in regional geographic characteristics. On the other hand, it can directly integrate information on changes in multiple land cover types and comprehensively consider the impact of changes in each type on the overall degree of change in the region.

[0039] The weights can be set according to the requirements of national geographic conditions monitoring and the importance level of land cover changes, or they can be obtained through historical change data analysis. For example, historical land cover type conversion data and the degree of impact on geographic information changes can be collected first, the correlation coefficient between land cover type conversion and the degree of geographic information change can be calculated, the contribution of land cover type conversion to the degree of change can be determined through regression analysis, and after normalization, the contribution can be converted into the corresponding weight value, and the total weight is 1, so as to accurately quantify the degree of land cover type conversion.

[0040] Extract the gray-level co-occurrence matrix contrast of candidate change regions from current and historical survey data, and calculate the absolute difference between the two as the texture feature variability.

[0041] The area change rate, land cover type conversion degree, and texture feature variability are normalized and weighted to calculate the change confidence of each candidate change region. The regions are then sorted in descending order of change confidence, and the top N candidate change regions are selected as each anomalous change patch, where N is the number determined proportionally based on the total number of candidate change regions.

[0042] It should be added that the normalized weighted fusion calculation formula for the change confidence is as follows: .

[0043] In the formula To vary the confidence level, , and These are the normalized area change rate, land cover type conversion degree, and texture feature variability, respectively. , and These are the weights for changes in the corresponding land cover types, used to quantify the differentiated impact of the three types of features on the confidence level of changes. For example, , , .

[0044] By calculating the confidence level of change through normalized weighted fusion, on the one hand, the weight allocation can reflect the different degrees of impact of area change, land cover type transformation, and texture feature variation on geographic information changes, reflecting the differences in their contributions to change judgment. On the other hand, it can unify features with different dimensions and physical meanings into a single evaluation system, comprehensively considering the combined impact of multi-dimensional information on the probability of change. The weights can be set according to geographic information update standards, regional land cover characteristics, or actual surveying requirements, or they can be obtained based on sample learning.

[0045] S2. Based on the preset change judgment rules, determine whether there are geographic information changes in the abnormal change patches. If there are, assess the urgency of the change based on the change confidence level and generate the change mapping priority for each abnormal change patch. If there are no changes, determine that the abnormal change patches are invalid changes and exclude them from the subsequent mapping path planning.

[0046] For example, determining whether there is a change in geographic information in the abnormally changed patch includes comparing the area change rate, the land cover type conversion degree, and the texture feature variation degree with their preset thresholds respectively.

[0047] If the land feature type conversion degree is greater than the preset type conversion threshold, it is determined that there is a change in geographic information.

[0048] If the land feature type conversion degree is less than or equal to the preset type conversion threshold, but the area change rate is greater than the preset area change threshold and the texture feature variability is greater than the preset texture variability threshold, then it is determined that there is a change in geographic information; otherwise, it is determined that there is no change in geographic information.

[0049] It should be added that the type conversion threshold is a threshold value used to determine whether the conversion of land cover types constitutes a significant change in geographic information. When the land cover type conversion degree exceeds this threshold, it indicates that a substantial change in land cover type has occurred. The area change threshold is a threshold value used to determine whether changes in regional area constitute a significant change in geographic information. When the area change rate exceeds this threshold, it indicates that a substantial change in spatial extent has occurred. The texture variation threshold is a threshold value used to determine whether changes in texture features constitute a significant change in geographic information. When the texture feature variation degree exceeds this threshold, it indicates that a substantial change in spatial structural features has occurred.

[0050] Taking the acquisition of the type conversion threshold as an example, the specific determination method includes the following steps: First, obtain samples of changed areas in historical surveying and mapping data that have been confirmed to have undergone changes in geographic information, and samples of non-changed areas that have been confirmed not to have undergone changes in geographic information.

[0051] Next, for each sample area, calculate its land cover type conversion degree, area change rate, and texture feature variability.

[0052] Then, the average value of the land cover type conversion degree of the changed area sample group and the unchanged area sample group is calculated as the type conversion threshold. Similarly, the area change threshold and texture variation threshold are calculated according to the calculation method of the type conversion threshold.

[0053] The transformation of land cover types directly reflects a fundamental change in land use and surface cover properties, and is the strongest evidence of substantial changes in geographic information. Therefore, when the degree of land cover type transformation exceeds a threshold on its own, the system considers it a decisive signal of structural or functional change, thus prioritizing and efficiently determining the existence of geographic information changes. When no significant land cover type transformation is detected, isolated area changes or texture variations may originate from non-substantial changes or data noise, such as seasonal growth or differences in surface moisture. Therefore, the rule requires that both the area change rate and texture feature variation exceed the threshold simultaneously for a change to be considered to exist. This logic constitutes a dual verification filter, effectively identifying and eliminating interference from variations in area without structural changes and texture disturbances without spatial expansion.

[0054] By assigning differentiated decision weights and combination logic to different characteristic indicators, a change determination system that is both efficient and robust was constructed. It prioritizes the most critical land cover type transformation while imposing stricter joint evidence requirements on other types of changes. This significantly reduces the false positive rate while ensuring a high recall rate, thus guaranteeing that the input information relied upon for subsequent mapping priority assessment and route planning has a high degree of accuracy and reliability.

[0055] For example, the priority of generating each abnormal change patch includes: summing the areas of each abnormal change patch to obtain the total area of ​​the abnormal change patches, and then using the ratio of the area of ​​each abnormal change patch to the total area as the area influence factor.

[0056] The change confidence level is multiplied by its area influence factor to obtain the change mapping influence of each abnormal change patch.

[0057] The abnormal change patches are sorted in descending order according to the change mapping impact, and the change mapping priority of each abnormal change patch is obtained. The higher the change mapping impact, the higher the priority.

[0058] S3. Based on the change mapping priority, the abnormal change patches are grouped into mapping partitions by spatial clustering algorithm, and a preliminary mapping task path is generated by sequence optimization algorithm. The spatial clustering algorithm includes, but is not limited to, the K-means algorithm or the DBSCAN algorithm, and the sequence optimization algorithm includes, but is not limited to, metaheuristic algorithms such as genetic algorithm and simulated annealing algorithm.

[0059] For example, the generation of the preliminary mapping task path includes: sorting each mapping partition in descending order according to priority to form a partition mapping sequence.

[0060] The abnormal change patches within the partition are sorted in descending order according to the degree of influence of the change in mapping, and a mapping sequence of patches within each partition is generated.

[0061] By connecting the mapping sequences of map features within each zone according to the zoning mapping sequence, a preliminary mapping task path is formed.

[0062] S4. Import the preliminary mapping task path into the geographic environment model to perform path safety conflict detection and multi-objective efficiency analysis. The geographic environment model is used to provide terrain, meteorological and airspace constraint information required for path planning.

[0063] For example, the path safety conflict detection includes: overlaying the preliminary mapping task path with the geographic environment model for analysis, and identifying path segments with flight altitudes below the safe ground clearance as terrain conflict segments.

[0064] The safe ground clearance is the minimum permissible altitude to ensure the safe flight of a mapping drone. This altitude is a comprehensive safety margin, and its specific value is determined primarily based on the aircraft's performance, the operational requirements of its onboard equipment, and airspace management regulations for the operating area. In practice, it can be set to a fixed value, such as 30 to 50 meters above the ground or ground obstacles. This altitude ensures that the aircraft can effectively avoid terrain undulations and potential obstacles, while meeting the basic requirements of mapping sensors for data acquisition quality.

[0065] Real-time wind speed data of the target area is acquired, and meteorological grids with wind speeds exceeding the operational threshold are extracted and recorded as risk meteorological grids. The risk meteorological grids are then matched with the spatiotemporal attributes of the preliminary mapping task path to identify meteorological conflict segments. The spatiotemporal attributes of the path include the three-dimensional spatial coordinates of each node and the estimated arrival time based on the average cruise speed.

[0066] It should be added that the specific steps for obtaining the meteorological conflict segment include: firstly, acquiring real-time meteorological radar data covering the target area, the data including gridded wind speed and wind direction information; performing quality control and spatial interpolation processing on the real-time meteorological radar data; and generating a meteorological element raster layer that is precisely registered with the geographic coordinate system.

[0067] Based on preset meteorological safety standards, risk meteorological grids are identified from the meteorological element grid layer, that is, grid cells with wind speed values ​​exceeding the safety threshold for drone operations are marked as risk meteorological grids.

[0068] Then, the preliminary mapping mission path is spatiotemporally overlaid with the risk weather grid: based on the spatial coordinates, flight elevation and expected passage time of each node on the path, it is matched with the spatial distribution of the risk weather grid and its corresponding effective time period to identify the segments in the path that intersect with the risk weather grid.

[0069] Finally, multiple intersecting segments that are spatially continuous or whose adjacent spacing is less than the preset merging threshold are merged into independent meteorological conflict segments, and the spatial range of each meteorological conflict segment is marked.

[0070] If no terrain or meteorological conflict sections are identified, it is determined that there are no safety conflicts in the preliminary surveying task path; otherwise, the overall safety risk coefficient is calculated by considering the spatial length and severity of each conflict section.

[0071] It should be added that the specific calculation process of the overall security risk coefficient is as follows: First, calculate the risk contribution value of each type of conflict segment identified.

[0072] The terrain conflict risk coefficient is calculated as follows: For each terrain conflict segment, the risk coefficient is jointly determined by the spatial length of the conflict segment and the severity of the terrain conflict. The severity is quantified as the difference between the aircraft's preset safe ground clearance and the actual terrain altitude. Therefore, the formula for calculating the terrain conflict risk coefficient for this segment is: .

[0073] In the formula For terrain conflict risk coefficient, The spatial length of the terrain conflict zone. To initially map the mission path length, For safe ground clearance, This represents the actual flight altitude. This formula shows that the longer the conflict segment and the greater the difference between the flight altitude and the safe altitude, the higher the risk contribution value.

[0074] The meteorological conflict risk coefficient is calculated as follows: For each meteorological conflict segment, the risk coefficient is jointly determined by the spatial length of the conflict segment and the severity of the meteorological conflict. The severity is quantified as the difference between the real-time wind speed and the safe operating wind speed threshold for the aircraft. Therefore, the formula for calculating the meteorological conflict risk coefficient for that segment is: .

[0075] In the formula For the meteorological conflict risk coefficient, The spatial length of the meteorological conflict zone. For real-time wind speed, This refers to the wind speed threshold for safe aircraft operation. This formula indicates that the longer the conflict zone and the greater the wind speed exceeding the safety threshold, the higher the risk contribution.

[0076] Finally, the risk coefficients of the terrain conflict section and the meteorological conflict section are normalized and summed to obtain the overall safety risk coefficient, which comprehensively reflects the overall level of terrain and meteorological safety risks faced on the preliminary mapping mission path.

[0077] When the overall security risk coefficient exceeds the benchmark security risk coefficient, a security conflict is determined to exist; otherwise, no security conflict is determined to exist.

[0078] The baseline safety risk coefficient refers to the overall safety risk coefficient calculated for the baseline path. The baseline path is the optimal traveling salesman loop (TSL) with the shortest total length, generated by a traveling salesman problem (TSP) algorithm, connecting the center points of all abnormally changed map features to be mapped. Its core function is to serve as an objective reference standard for evaluating the safety and efficiency of the initial mapping task path. It provides a safety benchmark for the subsequent quantitative comparison of the initial path safety risk coefficient, and also provides an efficiency benchmark for calculating path length efficiency, time efficiency, and overall efficiency. Specifically, the initially generated optimal TSL is imported into the geographic environment model, and a safety conflict detection process completely consistent with the initial mapping task path is executed. The TSL algorithm can employ metaheuristic algorithms such as genetic algorithms and simulated annealing algorithms to obtain an approximate optimal solution within a reasonable time.

[0079] If terrain or weather conflict sections are detected, the connection order of the center points of the abnormal change patches to be mapped needs to be adjusted based on the spatial location and extent of the conflict area. Then, a new optimal traveling salesman loop with the shortest total length is generated again using the traveling salesman problem algorithm. This process of generating a new optimal loop and detecting safety conflicts is repeated until the generated optimal traveling salesman loop does not identify any terrain or weather conflict sections, ensuring that the baseline path itself fully meets safety requirements.

[0080] For example, the multi-objective efficiency analysis includes: calculating the total length of the preliminary mapping task path and comparing it with the baseline path length to obtain the path length efficiency, wherein the baseline path length is the optimal traveling salesman loop length formed by connecting the center points of all the abnormal change patches to be mapped.

[0081] The preliminary mapping mission path was simulated, and the number of map features that were completely covered and the total number of map features with abnormal changes were counted during the simulated flight. The ratio of the two was used as the data acquisition efficiency.

[0082] The total mission execution time is estimated by dividing the total path length by the average cruise speed. The ratio of the baseline mission time to the total mission execution time is then used as the time efficiency. The baseline mission time is estimated by dividing the baseline path length by the average speed.

[0083] The overall efficiency is obtained by weighting and integrating path length efficiency, data acquisition efficiency, and time efficiency.

[0084] It should be added that the weighted fusion calculation formula for the overall efficiency is: Overall Efficiency .

[0085] In the formula For overall efficiency, , and These represent the normalized path length efficiency, data acquisition efficiency, and time efficiency, respectively. , and These are the corresponding weighting coefficients, used to quantify the differentiated impact of the three types of efficiency on overall efficiency, for example... , , .

[0086] By using weighted fusion to calculate overall efficiency, on the one hand, the weighted allocation can reflect the actual weights of the impact of path length efficiency, data acquisition efficiency, and time efficiency on different dimensions of the overall effectiveness of the surveying and mapping task, reflecting the differences in the contribution of different efficiency dimensions to the overall effectiveness. On the other hand, it can directly integrate efficiency information from the three dimensions of path length, data acquisition quality, and time cost, comprehensively considering the combined impact of the three on the overall execution effectiveness of the surveying and mapping task. The weights can be set based on the surveying and mapping task objectives, equipment performance requirements, or actual operational experience, or they can be obtained through analysis of historical task data.

[0087] The overall efficiency is compared with the baseline efficiency. If the overall efficiency is less than the baseline efficiency, the efficiency of the preliminary surveying and mapping task path is determined to be substandard; otherwise, the efficiency of the preliminary surveying and mapping task path is determined to be satisfactory. The baseline efficiency is the overall efficiency obtained by performing multi-objective efficiency analysis on the baseline path.

[0088] S5. Based on the detection and analysis results, determine whether the preliminary surveying task path needs to be corrected. If so, correct the preliminary surveying task path to obtain the final surveying task path; otherwise, use the preliminary surveying task path as the final surveying task path.

[0089] For example, determining whether the preliminary surveying task path needs to be modified includes: if it is determined that the preliminary surveying task path does not have security conflicts and meets the efficiency standards, then it is determined that the preliminary surveying task path does not need to be modified; otherwise, it is determined that the preliminary surveying task path needs to be modified.

[0090] For example, the modification of the preliminary surveying task path includes: if it is determined that there is a safety conflict in the preliminary surveying task path, then based on the location and range of the identified terrain conflict area or meteorological conflict section, taking the two ends of the conflict section as fixed start and end points, and taking improving the safety of the path as the main optimization goal, generating an avoidance path through a path replanning algorithm.

[0091] The conflict segments in the initial surveying task path are replaced with the corresponding avoidance paths to form a surveying task path with improved safety.

[0092] If the efficiency of the preliminary surveying task path is determined to be substandard, then the overall efficiency is used as the optimization objective function.

[0093] Using the current preliminary mapping task path as the initial solution, a metaheuristic algorithm is adopted as the sequence optimization algorithm. By iteratively adjusting the mapping partition sequence and the mapping sequence of abnormally changing patches in each partition, the task path that maximizes the value of the optimization objective function is searched to generate an efficiency-optimized mapping task path.

[0094] The mapping task path, after undergoing security conflict correction or efficiency optimization, is then imported back into the geographic environment model, and the path security conflict detection and multi-objective efficiency analysis are re-executed.

[0095] Repeat the above correction and verification process until the generated surveying task path is determined to have no security conflicts and meets the efficiency standards, and then use it as the final surveying task path.

[0096] Please see Figure 3 As shown, the present invention also provides a mapping task path planning system based on big data, which includes: a change patch identification module, a change priority generation module, a mapping path initial structure module, a path conflict detection module, and a task path decision module.

[0097] In the above, the change priority generation module is connected to the change patch recognition module and the mapping path initialization module, respectively, and the path conflict detection module is also connected to the mapping path initialization module and the task path decision module, respectively.

[0098] The change patch identification module acquires the current mapping data of the target area, compares it with historical mapping data to calculate the change confidence level, and identifies abnormal change patches accordingly.

[0099] The change priority generation module determines whether there are geographic information changes in abnormal change patches based on preset change judgment rules. If so, it assesses the urgency of the change based on the change confidence level and generates the change mapping priority for each abnormal change patch.

[0100] The initial mapping path module, based on the changed mapping priority, uses a spatial clustering algorithm to group abnormally changed patches into mapping partitions, and then uses a sequence optimization algorithm to generate a preliminary mapping task path.

[0101] The path conflict detection module imports the preliminary mapping task path into the geographic environment model to perform path safety conflict detection and multi-objective efficiency analysis.

[0102] The task path decision module determines whether the preliminary surveying task path needs to be corrected based on the detection and analysis results. If so, the preliminary surveying task path is corrected to obtain the final surveying task path; otherwise, the preliminary surveying task path is used as the final surveying task path.

[0103] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0104] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0105] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0106] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0107] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A mapping task path planning method based on big data, characterized in that: The method includes: S1. Obtain the current mapping data of the target area, compare it with the historical mapping data to calculate the change confidence level, and identify abnormal change patches accordingly; S2. Based on the preset change judgment rules, determine whether there are changes in geographic information in the abnormal change patches. If so, assess the urgency of the change based on the change confidence level and generate the change mapping priority for each abnormal change patch. S3. Based on the change mapping priority, the abnormal change patches are grouped into mapping partitions using a spatial clustering algorithm, and a preliminary mapping task path is generated using a sequence optimization algorithm accordingly. S4. Import the preliminary mapping task path into the geographic environment model and perform path safety conflict detection and multi-objective efficiency analysis. S5. Based on the detection and analysis results, determine whether the preliminary surveying task path needs to be corrected. If so, correct the preliminary surveying task path to obtain the final surveying task path; otherwise, use the preliminary surveying task path as the final surveying task path.

2. The mapping task path planning method based on big data according to claim 1, characterized in that: The identified abnormal change patches include: The near-infrared and red light band reflectance values ​​of each pixel are extracted from the current and historical mapping data, the normalized vegetation index is calculated based on the reflectance values, and the difference between the two normalized vegetation indices is calculated as the degree of difference. Calculate the average of the absolute values ​​of the differences of all pixels, and identify the set of pixels whose absolute values ​​of differences are greater than the average and form a continuous area in space as each candidate change area; Obtain the current area and historical area of ​​the candidate change region, and calculate their relative change rate as the area change rate; The area proportion of all dominant land cover types in the candidate change area is extracted from the current and historical survey data, the change rate of the area proportion of each dominant land cover type in the two time phases is calculated, and the area proportion change rate is weighted and fused to obtain the land cover type conversion degree. Extract the gray-level co-occurrence matrix contrast of candidate change regions from current and historical survey data, and calculate the absolute difference between the two as the texture feature variability. The area change rate, land cover type conversion degree, and texture feature variability are normalized and weighted to calculate the change confidence of each candidate change region. The regions are then sorted in descending order of change confidence, and the top N candidate change regions are selected as each anomalous change patch.

3. The mapping task path planning method based on big data according to claim 2, characterized in that: The determination of whether abnormally changed map features have undergone geographic information changes includes: The area change rate, land cover type conversion degree, and texture feature variability are compared with their preset thresholds respectively; If the land feature type conversion degree is greater than the preset type conversion threshold, it is determined that there is a change in geographic information; If the land feature type conversion degree is less than or equal to the preset type conversion threshold, but the area change rate is greater than the preset area change threshold and the texture feature variability is greater than the preset texture variability threshold, then it is determined that there is a change in geographic information; otherwise, it is determined that there is no change in geographic information.

4. The mapping task path planning method based on big data according to claim 1, characterized in that: The priority of generating change mapping for each abnormal change patch includes: The areas of each abnormal change patch are summed to obtain the total area of ​​the abnormal change patches. The ratio of the area of ​​each abnormal change patch to the total area is then used as the area influence factor. The change confidence level is multiplied by its area influence factor to obtain the change mapping influence of each abnormal change patch. The abnormal change patches are sorted in descending order according to the impact of change mapping to obtain the change mapping priority of each abnormal change patch.

5. The mapping task path planning method based on big data according to claim 1, characterized in that: The path for generating the preliminary mapping task includes: Sort each surveying and mapping zone in descending order according to priority to form a zone surveying and mapping sequence; The abnormal change patches within the partition are sorted in descending order according to the impact of the change in mapping, and a mapping sequence of patches within each partition is generated. By connecting the mapping sequences of map features within each zone according to the zoning mapping sequence, a preliminary mapping task path is formed.

6. The mapping task path planning method based on big data according to claim 1, characterized in that: The path security conflict detection includes: By overlaying the preliminary mapping mission path with the geographic environment model, path segments with flight altitudes below the safe ground clearance are identified as terrain conflict segments. Real-time wind speed data of the target area is acquired, and meteorological grids with wind speeds exceeding the operational threshold are extracted and recorded as risk meteorological grids. The risk meteorological grids are then matched with the spatiotemporal attributes of the preliminary mapping task path to identify meteorological conflict segments. If no terrain or meteorological conflict sections are identified, it is determined that there is no safety conflict in the preliminary surveying task path; otherwise, the overall safety risk coefficient is calculated by considering the spatial length and severity of each conflict section. When the overall security risk coefficient exceeds the benchmark security risk coefficient, a security conflict is determined to exist; otherwise, no security conflict is determined to exist.

7. The mapping task path planning method based on big data according to claim 1, characterized in that: The multi-objective efficiency analysis includes: Calculate the total length of the preliminary surveying task path and compare it with the baseline path length to obtain the path length efficiency. The preliminary mapping mission path was simulated, and the number of map patches that were completely covered and the total number of map patches with abnormal changes were counted during the simulated flight. The ratio of the two was used as the data acquisition efficiency. The total mission execution time is estimated by dividing the total path length by the average cruise speed, and the ratio of the baseline mission time to the total mission execution time is used as the time efficiency. The overall efficiency is obtained by weighted and fused calculation of path length efficiency, data acquisition efficiency and time efficiency. The overall efficiency is compared with the baseline efficiency. If the overall efficiency is less than the baseline efficiency, the efficiency of the preliminary surveying and mapping task path is determined to be substandard; otherwise, the efficiency of the preliminary surveying and mapping task path is determined to be up to standard.

8. The mapping task path planning method based on big data according to claim 1, characterized in that: The determination of whether the preliminary mapping task path needs to be corrected includes: If it is determined that the preliminary surveying task path does not have any security conflicts and meets the efficiency standards, then it is determined that the preliminary surveying task path does not need to be modified; otherwise, it is determined that the preliminary surveying task path needs to be modified.

9. The mapping task path planning method based on big data according to claim 1, characterized in that: The correction of the preliminary surveying task path includes: If it is determined that there is a safety conflict in the preliminary surveying task path, then based on the location and range of the identified terrain conflict area or meteorological conflict section, with the two ends of the conflict section as fixed start and end points, and with improving path safety as the main optimization goal, an avoidance path is generated through a path replanning algorithm. The conflict segments in the initial surveying task path are replaced with the corresponding avoidance paths to form a surveying task path with safety correction. If the efficiency of the preliminary surveying task path is determined to be substandard, then the overall efficiency will be used as the optimization objective function. Using the current preliminary mapping task path as the initial solution, a metaheuristic algorithm is adopted as the sequence optimization algorithm. By iteratively adjusting the mapping partition sequence and the mapping sequence of abnormally changed patches in each partition, the task path that maximizes the value of the optimization objective function is searched to generate an efficiency-optimized mapping task path. The mapping task path, after security conflict correction or efficiency optimization, is imported back into the geographic environment model, and the path security conflict detection and multi-objective efficiency analysis are re-executed. Repeat the above correction and verification process until the generated surveying task path is determined to have no security conflicts and meets the efficiency standards, and then use it as the final surveying task path.

10. A mapping task path planning system based on big data, characterized in that: The system includes: The change patch identification module acquires the current mapping data of the target area, compares it with historical mapping data to calculate the change confidence level, and identifies abnormal change patches accordingly. The change priority generation module determines whether there are geographic information changes in abnormal change patches based on preset change judgment rules. If so, it assesses the urgency of the change based on the change confidence level and generates the change mapping priority for each abnormal change patch. The initial mapping path module, based on the changed mapping priority, uses a spatial clustering algorithm to group abnormally changed patches to form mapping partitions, and then uses a sequence optimization algorithm to generate a preliminary mapping task path accordingly. The path conflict detection module imports the preliminary mapping task path into the geographic environment model and performs path safety conflict detection and multi-objective efficiency analysis. The task path decision module determines whether the preliminary surveying task path needs to be corrected based on the detection and analysis results. If so, the preliminary surveying task path is corrected to obtain the final surveying task path; otherwise, the preliminary surveying task path is used as the final surveying task path.

Images