Data fusion and optimization method in unmanned aerial vehicle surveying and mapping based on deep learning

Abstract

The invention discloses a data fusion and optimization method in unmanned aerial vehicle surveying and mapping based on deep learning, and relates to the technical field of unmanned aerial vehicle surveying and mapping control, and the method comprises the steps: dividing a to-be-surveyed and mapped region to obtain to-be-surveyed and mapped sub-regions, and collecting surveying and mapping complexity data of the to-be-surveyed and mapped sub-regions, the surveying and mapping complexity data comprising environmental data and geographic data; the surveying and mapping complexity data of the to-be-surveyed and mapped sub-region is processed to obtain a data fusion demand judgment result, a data fusion scheme is determined based on the data fusion demand judgment result, and the data fusion scheme is used for determining the data source fusion number of data fusion; and performing data fusion on the surveying and mapping data of the to-be-surveyed and mapped sub-region according to the data fusion scheme The problem that large-area surveying and mapping possibly face too large regional environment difference is solved, data fusion in unnecessary regions is avoided, the system load is effectively reduced, and the efficiency and pertinence of unmanned aerial vehicle surveying and mapping data processing are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV mapping control, and specifically to a data fusion and optimization method in UAV mapping based on deep learning. Background Art

[0002] With the rapid development of related technologies, the sensors carried by UAVs have become more accurate and diverse, endowing UAVs with powerful sensing capabilities, and also making the processing and analysis of multi-sensor data a major challenge in UAV applications. Data fusion is a key technology to solve this problem, which realizes the fusion and utilization of multi-sensor data through processes such as detection, association, combination, and estimation, and obtains accurate UAV states and target information to provide support for decision-making.

[0003] For example, the invention patent with the publication number CN119045511B discloses a mapping UAV and a mapping method for reducing mapping errors. The present invention uses the radar device and camera of the UAV to monitor the surrounding environment in real time, determines the position of obstacles through data fusion technology and image recognition algorithms, establishes an obstacle avoidance decision-making model according to the target mapping task, introduces a simulated annealing algorithm to optimize the model and formulate an obstacle avoidance strategy, and finally feeds back the obstacle avoidance strategy to the control center. The control center conducts route planning and real-time controls the attitude adjustment speed of the UAV to obtain the final route.

[0004] For example, the invention patent with the publication number CN117519269A discloses a UAV system for forestry survey and mapping, including the UAV system for forestry survey and mapping, which is used for forest land monitoring, resource assessment, pest detection, fire monitoring, terrain survey, image acquisition, road pipeline monitoring, and wildlife monitoring; a sensing integration module that integrates multi-sensing units and their hardware devices to ensure that they can work together and communicate with the UAV system; a data acquisition and synchronization module that synchronizes data into the UAV system by providing a unified data acquisition interface; a data processing and fusion module that establishes data processing processes and algorithms to fuse data from different sensing units. The hydrological sensing unit, multi-spectral sensing unit, meteorological sensing unit, and hardware integration unit in the sensing integration module meet more survey requirements and avoid the limited performance and resolution of sensors.

[0005] However, in the process of implementing the technical solutions of the present invention in the embodiments of the present application, it is found that the above technologies have at least the following technical problems: in the scenario of large-area mapping, real-time data fusion will increase the system load, and at the same time, large-area mapping may face the problem of too large regional environmental differences, and the degree of fuzzy error of information from different data sources is different. Summary of the Invention

[0006] In view of the deficiencies of the prior art, the present invention provides a data fusion and optimization method in UAV mapping based on deep learning, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions: In a first aspect of the present invention, a data fusion and optimization method in UAV mapping based on deep learning is provided, including: dividing the area to be mapped to obtain sub-areas to be mapped, and collecting mapping complexity data of the sub-areas to be mapped, where the mapping complexity data includes environmental data and geographical data.

[0008] Processing the mapping complexity data of the sub-areas to be mapped to obtain a data fusion requirement determination result, and determining a data fusion scheme based on the data fusion requirement determination result, where the data fusion scheme is used to determine the number of data source fusions for data fusion.

[0009] Performing data fusion on the mapping data of the sub-areas to be mapped according to the data fusion scheme.

[0010] As a further method, the geographical data includes elevation, slope, terrain undulation degree, and surface roughness, and the environmental data includes light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity; the process of specifically analyzing the processing of the mapping complexity data of the sub-areas to be mapped to obtain a data fusion requirement determination result is as follows: extracting preset mapping complexity data reference values from the UAV mapping database, where the mapping complexity data reference values include critical elevation, critical slope, critical terrain undulation degree, critical surface roughness, reference light intensity, allowable light intensity deviation value, reference humidity, allowable humidity deviation value, critical wind speed, critical visibility, and critical electromagnetic interference intensity; processing the geographical data and environmental data respectively to obtain the terrain complexity and environmental complexity of the sub-areas to be mapped; processing the terrain complexity and environmental complexity to obtain a data fusion requirement evaluation value for the sub-areas to be mapped; comparing the data fusion requirement evaluation value with a preset data fusion requirement evaluation threshold in the UAV mapping database: if the data fusion requirement evaluation value is less than the preset data fusion requirement evaluation threshold, then record the data fusion requirement determination result of the sub-areas to be mapped as not performing data fusion; if the data fusion requirement evaluation value is greater than or equal to the preset data fusion requirement evaluation threshold, then record the data fusion requirement determination result of the sub-areas to be mapped as performing data fusion and mark it as a sub-area to be fused, and determine the number of data source fusions according to the data fusion requirement evaluation value.

[0011] As a further method, the terrain complexity and environmental complexity of the sub-region to be surveyed and mapped are respectively processed based on geographical data and environmental data. The specific analysis process is as follows: The elevation influence degree is obtained by performing an elevation ratio approaching influence degree operation on the elevation data of the sub-region to be surveyed and mapped and a preset critical elevation. The elevation ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree between the elevation data of the sub-region to be surveyed and mapped and the preset critical elevation corrected by a preset elevation weight factor; The slope influence degree is obtained by performing a slope ratio approaching influence degree operation on the slope data of the sub-region to be surveyed and mapped and a preset critical slope. The slope ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree between the slope data of the sub-region to be surveyed and mapped and the preset critical slope corrected by a preset slope weight factor; The terrain undulation influence degree is obtained by performing a terrain undulation ratio approaching influence degree operation on the terrain undulation data of the sub-region to be surveyed and mapped and a preset critical terrain undulation. The terrain undulation ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree between the terrain undulation data of the sub-region to be surveyed and mapped and the preset critical terrain undulation corrected by a preset terrain undulation weight factor; The surface roughness influence degree is obtained by performing a surface roughness ratio approaching influence degree operation on the surface roughness data of the sub-region to be surveyed and mapped and a preset critical surface roughness. The surface roughness ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree between the surface roughness data of the sub-region to be surveyed and mapped and the preset critical surface roughness corrected by a preset surface roughness weight factor; The terrain complexity of the sub-region to be surveyed and mapped is obtained by superimposing the elevation influence degree, slope influence degree, terrain undulation influence degree and surface roughness influence degree. The terrain complexity represents the quantitative data of the complexity degree of elevation, slope, terrain undulation and surface roughness for the geographical factors on the sub-region to be surveyed and mapped; The illumination influence degree is obtained by performing an illumination intensity allowable deviation correction operation on the illumination intensity data of the sub-region to be surveyed and mapped and a preset reference illumination intensity. The illumination intensity allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree between the illumination intensity data of the sub-region to be surveyed and mapped and the preset reference illumination intensity corrected by a preset illumination weight factor; The humidity influence degree is obtained by performing a humidity allowable deviation correction operation on the humidity data of the sub-region to be surveyed and mapped and a preset reference humidity. The humidity allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree between the humidity data of the sub-region to be surveyed and mapped and the preset reference humidity corrected by a preset humidity weight factor; The wind speed influence degree is obtained by performing a wind speed ratio approaching influence degree operation on the wind speed data of the sub-region to be surveyed and mapped and a preset critical wind speed. The wind speed ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree between the wind speed data of the sub-region to be surveyed and mapped and the preset critical wind speed corrected by a preset wind speed weight factor;The visibility impact degree is obtained by performing a visibility ratio approaching impact degree operation on the visibility data of the sub-region to be surveyed and the preset critical visibility. The visibility ratio approaching impact degree operation represents the quantitative data of the complexity impact corresponding to the approaching degree of the visibility data of the sub-region to be surveyed and the preset critical visibility corrected by the preset visibility weight factor. The electromagnetic interference impact degree is obtained by performing an electromagnetic interference intensity ratio approaching impact degree operation on the electromagnetic interference intensity data of the sub-region to be surveyed and the preset critical electromagnetic interference intensity. The electromagnetic interference intensity ratio approaching impact degree operation represents the quantitative data of the complexity impact corresponding to the approaching degree of the electromagnetic interference intensity data of the sub-region to be surveyed and the preset critical electromagnetic interference intensity corrected by the preset electromagnetic interference weight factor. The environmental complexity of the sub-region to be surveyed is obtained by superimposing the light impact degree, humidity impact degree, wind speed impact degree, visibility impact degree, and electromagnetic interference impact degree. The environmental complexity represents the quantitative data of the complexity of the environmental factors such as light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity on the sub-region to be surveyed.

[0012] As a further method, the data fusion requirement evaluation value of the sub-region to be surveyed is obtained by processing the terrain complexity and the environmental complexity. The specific analysis process is as follows: The terrain impact degree is obtained by performing a weight assignment process on the terrain complexity, the environmental impact degree is obtained by performing a weight assignment process on the environmental complexity, and the data fusion requirement evaluation value of the sub-region to be surveyed is obtained by superimposing the terrain impact degree and the environmental impact degree. The data fusion requirement evaluation value represents the quantitative data of the necessity of the terrain complexity and the environmental complexity for data fusion in the UAV surveying of the sub-region to be surveyed.

[0013] As a further method, the number of data source fusions is determined according to the data fusion requirement evaluation value. The specific analysis process is as follows: The data fusion requirement evaluation value is matched with the preset data fusion requirement level interval in the UAV surveying database to obtain the number of data source fusions corresponding to the preset data fusion requirement level interval required for the sub-region to be surveyed. The parameter adjustment determination value is obtained by subtracting the data fusion requirement evaluation value from the data fusion requirement threshold. The parameter adjustment determination value is compared with the preset sensor parameter adjustment threshold in the UAV surveying database: If the parameter adjustment determination value is greater than or equal to the preset sensor parameter adjustment threshold, the sensor parameters are adjusted; if the parameter adjustment determination value is less than the preset sensor parameter adjustment threshold, no additional operation is performed.

[0014] As a further method, for the adjustment of the sensor parameters, the specific analysis process is as follows: Obtain the preset terrain complexity threshold and environment complexity threshold from the UAV mapping database. By subtracting the terrain complexity from the terrain complexity threshold, the terrain deviation degree is obtained; by subtracting the environment complexity from the environment complexity threshold, the environment deviation degree is obtained; by matching the terrain deviation degree, the environment deviation degree with the preset parameter adjustment strategy table in the UAV mapping database to obtain the specific adjustment amount, and adjusting the sensitivity, resolution and sampling frequency of the sensor.

[0015] As a further method, for the data fusion of the mapping data of the sub-region to be mapped according to the data fusion scheme, the specific analysis process is as follows: Obtain the UAV operation data, where the UAV operation data includes the UAV flight altitude, flight speed and flight line overlap rate; Extract the preset UAV operation data reference values from the UAV mapping database, where the UAV operation data reference values include the reference flight altitude, allowable flight altitude deviation, reference flight speed, allowable flight speed deviation, reference flight line overlap rate and allowable flight line overlap rate deviation; By performing a flight altitude allowable deviation correction operation on the flight altitude data of the sub-region to be fused with the preset reference flight altitude, the flight altitude influence degree is obtained, and the flight altitude allowable deviation correction operation represents the quantization data of the complexity influence corresponding to the deviation degree of the flight altitude data of the sub-region to be fused with the preset reference flight altitude corrected by the preset flight altitude weight factor; By performing a flight speed allowable deviation correction operation on the flight speed data of the sub-region to be fused with the preset reference flight speed, the flight speed influence degree is obtained, and the flight speed allowable deviation correction operation represents the quantization data of the complexity influence corresponding to the deviation degree of the flight speed data of the sub-region to be fused with the preset reference flight speed corrected by the preset flight speed weight factor; By performing a flight line overlap rate allowable deviation correction operation on the flight line overlap rate data of the sub-region to be fused with the preset reference flight line overlap rate, the flight line overlap rate influence degree is obtained, and the flight line overlap rate allowable deviation correction operation represents the quantization data of the complexity influence corresponding to the deviation degree of the flight line overlap rate data of the sub-region to be mapped with the preset reference flight line overlap rate corrected by the preset flight line overlap rate weight factor; By superimposing the flight altitude influence degree, the flight speed influence degree and the flight line overlap rate influence degree, the operation state value is obtained; By performing a weighting process in combination with the terrain complexity, environment complexity and operation state value, the data source accuracy influence evaluation value of the sensor in the sub-region to be fused is obtained, and the data source accuracy influence evaluation value represents the quantization data of the influence degree of the terrain complexity, environment complexity, flight altitude, flight speed and flight line overlap rate on the quality of the data source of a certain sensor in the current sub-region to be fused; Sort according to the data source accuracy influence evaluation value to obtain the data sources for data fusion.

[0016] As a further method, the data sources are sorted according to the data source accuracy impact assessment value, and the data source is obtained for data fusion. The specific analysis process is: sort the data source accuracy impact assessment value of each sensor from low to high; compare the data source accuracy impact assessment value with the data source accuracy impact assessment threshold preset in the UAV mapping database, count the number of data sources that are less than or equal to the data source accuracy impact assessment threshold, and mark them as the number of candidate data sources that can be fused; compare the number of candidate data sources that can be fused with the number of data source fusions, if the number of candidate data sources that can be fused is greater than or equal to the number of data source fusions, select the data source for data fusion according to the data source accuracy impact assessment value from low to high; if the number of candidate data sources that can be fused is less than the number of data source fusions, adjust the UAV parameters and sensor parameters, and perform secondary collection; if the collection time is insufficient, reduce the data source accuracy requirements and increase the number of data source fusions.

[0017] As a further method, the adjustment of the UAV parameters and sensor parameters, the specific analysis process is: by subtracting the number of candidate data sources that can be fused from the number of data source fusions, the data source supply and demand difference is obtained; the data source supply and demand difference is compared with the difference adjustment threshold interval preset in the UAV surveying and mapping database, if the data source supply and demand difference is in the low difference adjustment threshold interval, only the sensor parameters are adjusted; if the data source supply and demand difference is in the high difference adjustment threshold interval, the UAV parameters and sensor parameters are adjusted synchronously; according to the data source supply and demand difference and the sensor parameter adjustment range and the UAV parameter adjustment range preset in the UAV surveying and mapping database, the specific adjustment amount is matched, and the sensor sensitivity, resolution and sampling frequency are adjusted with the UAV's flight altitude, flight speed and route overlap rate.

[0018] As a further method, the data source accuracy requirement is reduced and the number of data sources to be fused is increased. The specific analysis process is as follows: the data source accuracy impact assessment mean of the sub-area to be fused is calculated based on the data source accuracy impact assessment values ​​of all sensors in the sub-area to be fused, the data source accuracy impact assessment mean is subtracted from a preset data source accuracy impact assessment threshold, and the data source accuracy impact assessment threshold is corrected based on the difference; the data source accuracy impact assessment value is compared with the data source accuracy impact assessment threshold after the correction, a data source that is less than or equal to the data source accuracy impact assessment threshold after the correction is obtained, and a high-level fusion algorithm is started for data fusion.

[0019] Compared with the prior art, the embodiments of the present invention have at least the following beneficial effects:

[0020] (1) By collecting the geographical data and environmental data of the sub-region to be surveyed and mapped, comprehensively analyzing to obtain the terrain complexity and environmental complexity, and then determining the data fusion requirement evaluation value, it is judged whether to perform data fusion and the number of data source fusions required according to the evaluation value, avoiding data fusion in unnecessary regions, effectively reducing the system load, and improving the efficiency and pertinence of UAV surveying and mapping data processing.

[0021] (2) By combining the terrain complexity, environmental complexity and UAV operation data, the present invention calculates the data source accuracy impact evaluation value, sorts and filters the data sources accordingly, and selects appropriate data sources for fusion, fully considering the influence of various factors on the quality of data sources, improving the accuracy and reliability of data fusion, and making the final surveying and mapping results better reflect the actual terrain and environmental conditions.

[0022] (3) By adjusting the UAV parameters and sensor parameters according to the difference between the supply and demand of data sources, and reasonably reducing the data source accuracy requirements and increasing the number of data source fusions when the acquisition time is insufficient, it is ensured that surveying and mapping data meeting certain requirements can be obtained under different surveying and mapping conditions, enhancing the adaptability and flexibility of the UAV surveying and mapping system, and expanding the application scenarios of UAV surveying and mapping.

[0023] (4) By combining the terrain deviation degree and environmental deviation degree, and adjusting the sensitivity, resolution and sampling frequency of the sensor by referring to the preset parameter adjustment strategy table, the sensor can better adapt to the complex terrain and environmental conditions of different surveying and mapping regions. It can not only ensure the data quality, but also reduce data redundancy, lower the data processing cost, and effectively improve the data acquisition performance of the sensor in various surveying and mapping scenarios.

[0024] Of course, any product implementing the present invention does not necessarily need to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0028] Referring Figure 1 As shown, the first aspect of the present invention provides a data fusion and optimization method in drone mapping based on deep learning, including: dividing the area to be mapped to obtain sub-areas to be mapped, and collecting mapping complexity data of the sub-areas to be mapped, where the mapping complexity data includes environmental data and geographical data.

[0029] Processing the mapping complexity data of the sub-areas to be mapped to obtain a data fusion requirement determination result, and determining a data fusion scheme based on the data fusion requirement determination result, where the data fusion scheme is used to determine the number of data sources to be fused in the data fusion.

[0030] Specifically, the geographical data includes elevation, slope, terrain undulation degree, and surface roughness, and the environmental data includes light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity.

[0031] In this embodiment, in the urban three-dimensional modeling task, the drone divides the area to be mapped, collects the geographical data and environmental data of each sub-area, generates a data fusion requirement determination result, and determines the number of data sources to be fused. For example, in a complex building area (high terrain undulation degree, low visibility), multi-sensor fusion is used to ensure high-precision modeling; while in a flat area (low slope, high light intensity), only a single sensor is required to meet the requirements. The area division adopts the grid division method. According to the overall scope and shape of the area to be mapped, it is divided into uniformly sized square or rectangular grids, and each grid is a sub-area to be mapped, which is convenient for management and data collection. The size of the grid can be adjusted according to the actual mapping requirements and accuracy requirements.

[0032] It should be understood that in this embodiment, the elevation refers to the vertical distance from a ground point to the geoid, which can be obtained through a global navigation satellite system, such as GPS (Global Positioning System) or the Beidou satellite navigation system. The slope refers to the degree of steepness of a surface unit and can be obtained by using a slope extraction algorithm in geographic information system software through a digital elevation model. The terrain undulation degree refers to the difference in altitude between the highest and lowest points in a region, which is used to measure the undulation degree of the terrain and can be obtained by analyzing digital elevation model data. The surface roughness refers to the roughness and irregularity of the surface and can be measured by a synthetic aperture radar. The light intensity refers to the luminous flux of visible light received per unit area and can be measured by a light sensor. The humidity refers to the water vapor content in the air and can be measured by a humidity sensor. The wind speed refers to the speed of air flow, which affects the flight stability and mapping accuracy of an unmanned aerial vehicle and can be measured by a wind speed sensor. The visibility refers to the transparency of the atmosphere, which affects the effective detection distance of an optical sensor and can be measured by a visibility meter. The electromagnetic interference intensity is used to measure the strength of electromagnetic interference in space and can be obtained by an electromagnetic interference measuring instrument.

[0033] Specifically, the data fusion requirement determination result is obtained by processing the mapping complexity data of the sub-region to be mapped. The specific analysis process is as follows: Extract the preset mapping complexity data reference values from the unmanned aerial vehicle mapping database. The mapping complexity data reference values include the critical elevation, critical slope, critical terrain undulation degree, critical surface roughness, reference light intensity, allowable light intensity deviation value, reference humidity, allowable humidity deviation value, critical wind speed, critical visibility, and critical electromagnetic interference intensity. Process the geographic data and environmental data respectively to obtain the terrain complexity and environmental complexity of the sub-region to be mapped. Process the terrain complexity and environmental complexity to obtain the data fusion requirement evaluation value of the sub-region to be mapped. Compare the data fusion requirement evaluation value with the preset data fusion requirement evaluation threshold in the unmanned aerial vehicle mapping database: If the data fusion requirement evaluation value is less than the preset data fusion requirement evaluation threshold, record the data fusion requirement determination result of the sub-region to be mapped as not performing data fusion; if the data fusion requirement evaluation value is greater than or equal to the preset data fusion requirement evaluation threshold, record the data fusion requirement determination result of the sub-region to be mapped as performing data fusion and mark it as a sub-region to be fused, and determine the number of data source fusions according to the data fusion requirement evaluation value.

[0034] Specifically, the terrain complexity and environmental complexity of the sub-region to be surveyed and mapped are obtained by processing geographical data and environmental data respectively. The specific analysis process is as follows: the elevation influence degree is obtained by performing an elevation ratio approaching influence degree operation on the elevation data of the sub-region to be surveyed and mapped and a preset critical elevation. The elevation ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the elevation data of the sub-region to be surveyed and mapped and the preset critical elevation corrected by a preset elevation weight factor; the slope influence degree is obtained by performing a slope ratio approaching influence degree operation on the slope data of the sub-region to be surveyed and mapped and a preset critical slope. The slope ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the slope data of the sub-region to be surveyed and mapped and the preset critical slope corrected by a preset slope weight factor; the terrain undulation influence degree is obtained by performing a terrain undulation ratio approaching influence degree operation on the terrain undulation data of the sub-region to be surveyed and mapped and a preset critical terrain undulation. The terrain undulation ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the terrain undulation data of the sub-region to be surveyed and mapped and the preset critical terrain undulation corrected by a preset terrain undulation weight factor; the surface roughness influence degree is obtained by performing a surface roughness ratio approaching influence degree operation on the surface roughness data of the sub-region to be surveyed and mapped and a preset critical surface roughness. The surface roughness ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the surface roughness data of the sub-region to be surveyed and mapped and the preset critical surface roughness corrected by a preset surface roughness weight factor; the terrain complexity of the sub-region to be surveyed and mapped is obtained by superimposing the elevation influence degree, the slope influence degree, the terrain undulation influence degree and the surface roughness influence degree. The terrain complexity represents the quantitative data of the complexity of the elevation, slope, terrain undulation and surface roughness for the geographical factors on the sub-region to be surveyed and mapped; the light influence degree is obtained by performing a light intensity allowable deviation correction operation on the light intensity data of the sub-region to be surveyed and mapped and a preset reference light intensity. The light intensity allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree of the light intensity data of the sub-region to be surveyed and mapped and the preset reference light intensity corrected by a preset light weight factor; the humidity influence degree is obtained by performing a humidity allowable deviation correction operation on the humidity data of the sub-region to be surveyed and mapped and a preset reference humidity. The humidity allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree of the humidity data of the sub-region to be surveyed and mapped and the preset reference humidity corrected by a preset humidity weight factor; the wind speed influence degree is obtained by performing a wind speed ratio approaching influence degree operation on the wind speed data of the sub-region to be surveyed and mapped and a preset critical wind speed. The wind speed ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the wind speed data of the sub-region to be surveyed and mapped and the preset critical wind speed corrected by a preset wind speed weight factor;The visibility influence degree is obtained by performing a visibility ratio approaching influence degree operation on the visibility data of the sub-region to be surveyed and mapped and a preset critical visibility. The visibility ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the visibility data of the sub-region to be surveyed and mapped and the preset critical visibility corrected by a preset visibility weight factor. The electromagnetic interference influence degree is obtained by performing an electromagnetic interference intensity ratio approaching influence degree operation on the electromagnetic interference intensity data of the sub-region to be surveyed and mapped and a preset critical electromagnetic interference intensity. The electromagnetic interference intensity ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the electromagnetic interference intensity data of the sub-region to be surveyed and mapped and the preset critical electromagnetic interference intensity corrected by a preset electromagnetic interference weight factor. The environmental complexity of the sub-region to be surveyed and mapped is obtained by superimposing the light influence degree, humidity influence degree, wind speed influence degree, visibility influence degree, and electromagnetic interference influence degree. The environmental complexity represents the quantitative data of the complexity degree of the environmental factors on the sub-region to be surveyed and mapped by light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity.

[0035] In a specific embodiment, the ways to obtain the terrain complexity and environmental complexity are as follows:

[0036]

[0037] α1 + α2 + α3 + α4 = 1;

[0038] β1 + β2 + β3 + β4 + β5 = 1;

[0039] In the formula, DX i represents the terrain complexity of the i-th sub-region to be surveyed and mapped, HJ i represents the environmental complexity of the i-th sub-region to be surveyed and mapped, GC i represents the elevation of the i-th sub-region to be surveyed and mapped, GC0 represents the preset critical elevation, PD i represents the slope of the i-th sub-region to be surveyed and mapped, PD0 represents the preset critical slope, DQ i represents the terrain undulation of the i-th sub-region to be surveyed and mapped, DQ0 represents the preset critical terrain undulation, DC i represents the surface roughness of the i-th sub-region to be surveyed and mapped, DC0 represents the preset critical surface roughness, GQ i represents the light intensity of the i-th sub-region to be surveyed and mapped, represents the preset reference light intensity, ΔGQ represents the preset allowable light intensity deviation value, SD i represents the humidity of the i-th sub-region to be surveyed and mapped, represents the preset reference humidity, ΔSD represents the preset allowable humidity deviation value, FS iThe wind speed of the i-th sub-region to be surveyed and mapped, FS0 represents the preset critical wind speed, ND i The visibility of the i-th sub-region to be surveyed and mapped, ND0 represents the preset critical visibility, DG i The electromagnetic interference intensity of the i-th sub-region to be surveyed and mapped, DG0 represents the preset critical electromagnetic interference intensity, α1 represents the preset elevation weight factor, α2 represents the preset slope weight factor, α3 represents the preset terrain undulation weight factor, α4 represents the preset surface roughness weight factor, β1 represents the preset light intensity weight factor, β2 represents the preset humidity weight factor, β represents the preset wind speed weight factor, β4 represents the preset visibility weight factor, β5 represents the preset electromagnetic interference intensity weight factor, i represents the number of the sub-region to be surveyed and mapped, i = 1, 2, 3,..., k, and k represents the total number of sub-regions to be surveyed and mapped.

[0040] When using α1, α2, α3, and α4, the weight factors corresponding to the terrain complexity of elevation, slope, terrain undulation, and surface roughness can be directly obtained from the UAV surveying and mapping database, and respectively represent the numerical values of the influence degrees of elevation, slope, terrain undulation, and surface roughness on the terrain complexity. The elevation weight factor, slope weight factor, terrain undulation weight factor, and surface roughness weight factor and elevation, slope, terrain undulation, and surface roughness can be a preset mapping relationship. For example, inputting elevation, slope, terrain undulation, and surface roughness into the corresponding preset mapping sets respectively to obtain the weight factors in the terrain complexity process corresponding to elevation, slope, terrain undulation, and surface roughness, and the mapping relationship therein can be one-to-one or many-to-one. In this embodiment, the value ranges of the elevation weight factor, slope weight factor, terrain undulation weight factor, and surface roughness factor are all limited between 0 and 1, and the sum of the elevation weight factor, slope weight factor, terrain undulation weight factor, and surface roughness weight factor is 1.

[0041] When β1, β2, β3, β4, and β5 are used, the weight factors corresponding to the environmental complexity of light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity can be directly obtained from the UAV mapping database, and they respectively represent the numerical values of the influence degrees of light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity on the environmental complexity. The mapping relationships between the light intensity weight factor, humidity weight factor, wind speed weight factor, visibility weight factor, and electromagnetic interference intensity weight factor and light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity can be preset. For example, by inputting light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity into the corresponding preset mapping sets respectively, the weight factors corresponding to the environmental complexity of light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity are obtained. The mapping relationships therein can be one-to-one or many-to-one relationships. In this embodiment, the value ranges of the light intensity weight factor, humidity weight factor, wind speed weight factor, visibility weight factor, and electromagnetic interference intensity weight factor are all limited between 0 and 1, and the sum of the light intensity weight factor, humidity weight factor, wind speed weight factor, visibility weight factor, and electromagnetic interference intensity weight factor is 1.

[0042] In this embodiment, the terrain complexity is used to quantitatively evaluate the complexity degree of geographical factors on the sub-region to be mapped. The greater the elevation, or the greater the slope, or the greater the terrain undulation, or the greater the surface roughness, the greater the terrain complexity, indicating that the geographical factors have a higher complexity degree on the sub-region to be mapped.

[0043] In this embodiment, the environmental complexity is used to quantitatively evaluate the complexity degree of environmental factors on the sub-region to be mapped. The greater the deviation between the light intensity and the reference light intensity, or the greater the deviation between the humidity and the reference humidity, or the faster the wind speed, or the lower the visibility, or the stronger the electromagnetic interference intensity, the higher the environmental complexity, indicating that the environmental factors have a higher complexity degree on the sub-region to be mapped.

[0044] The algorithm of this embodiment combines elevation, slope, terrain undulation and surface roughness, and comprehensively analyzes to obtain terrain complexity. In this formula, elevation, slope, terrain undulation and surface roughness affect each other. As the elevation increases, the slope also increases accordingly. For example, in mountainous areas, the inclination of the mountain will also increase as the terrain continues to rise. Because the change in elevation changes the stress conditions of the earth's crust, it affects the inclination of the surface. At the same time, the increase in elevation will also lead to an increase in terrain undulation. As the altitude increases, the difference between the peak and the valley is also increasing. The increase in terrain undulation will increase the surface roughness, because in places with large terrain differences, rock crushing, weathering and other effects are more intense, and the surface will become rougher. As the slope increases, the terrain undulation will also increase. In some mountainous areas, the terrain difference associated with steep slopes is also greater. In areas with large slopes, surface materials are more likely to move under the influence of factors such as gravity, resulting in increased surface roughness. For example, in some steep slope areas of the Loess Plateau, soil erosion is serious, the surface is bumpy, and the surface roughness increases. By comprehensively analyzing elevation, slope, terrain undulation, and surface roughness, the terrain complexity can be accurately obtained, which quantitatively reflects the terrain complexity of the sub-area to be surveyed and mapped due to geographical factors during the surveying process and the degree of deviation from the ideal surveying terrain conditions.

[0045] The algorithm of this embodiment combines light intensity, humidity, wind speed, visibility and electromagnetic interference intensity, and comprehensively analyzes to obtain environmental complexity. In this formula, light intensity, humidity, wind speed, visibility and electromagnetic interference intensity affect each other. As the deviation between light intensity and reference light intensity increases, visibility will decrease. For example, in the early morning or evening, the light intensity becomes weaker, and the deviation from the reference light intensity during normal sunlight becomes larger. At this time, particles in the air are more likely to scatter light, resulting in reduced visibility. At the same time, changes in light intensity will also affect humidity. When the light becomes weaker, the ground temperature decreases, and water vapor is more likely to condense, thereby increasing humidity. As the deviation between humidity and reference humidity increases, the electromagnetic interference intensity will increase. In a humid environment, moisture will cause a conductive water film to form on the surface of some objects, interfering with the propagation of electromagnetic signals, resulting in an increase in electromagnetic interference intensity. In addition, increased humidity will also cause fog to form, further reducing visibility. As the wind speed increases, the dust and particulate matter in the air will increase, thereby reducing visibility. For example, in sandstorm weather, strong winds carry a lot of sand and dust, causing visibility to drop sharply. At the same time, changes in wind speed will also affect humidity. Strong winds will accelerate the evaporation of water or blow away water vapor, changing the humidity conditions in local areas. By comprehensively analyzing light intensity, humidity, wind speed, visibility and electromagnetic interference intensity, the environmental complexity can be accurately obtained, which quantitatively reflects the environmental complexity of the sub-area to be surveyed during the surveying process and the degree of deviation from the ideal surveying environment conditions.

[0046] Specifically, the data fusion requirement evaluation value of the sub-region to be surveyed and mapped is obtained according to the terrain complexity and environmental complexity. The specific analysis process is as follows: the terrain influence degree is obtained by weighting the terrain complexity, the environmental influence degree is obtained by weighting the environmental complexity, and the data fusion requirement evaluation value of the sub-region to be surveyed and mapped is obtained by superimposing the terrain influence degree and the environmental influence degree; the data fusion requirement evaluation value represents the quantitative data of the necessary degree of the terrain complexity and environmental complexity for data fusion in the UAV surveying and mapping of the sub-region to be surveyed and mapped.

[0047] In a specific embodiment, the acquisition method of the data fusion requirement evaluation value is as follows:

[0048] SX i = DX i *γ1 + HJ i *γ2;

[0049] γ1 + γ2 = 1;

[0050] In the formula, SX i represents the data fusion requirement evaluation value of the i-th sub-region to be surveyed and mapped, DX i represents the terrain complexity of the i-th sub-region to be surveyed and mapped, HJ i represents the environmental complexity of the i-th sub-region to be surveyed and mapped, γ represents the preset terrain complexity weight factor, γ2 represents the preset environmental complexity weight factor, i represents the number of the sub-region to be surveyed and mapped, i = 1, 2, 3,..., k, and k represents the total number of sub-regions to be surveyed and mapped.

[0051] γ1 and γ2 can directly obtain the weight factors in the process of obtaining the data fusion requirement evaluation values corresponding to the terrain complexity and environmental complexity from the UAV surveying and mapping database, and respectively represent the numerical values of the influence degrees of the terrain complexity and environmental complexity on the data fusion requirement evaluation value. The terrain complexity weight factor and the environmental complexity weight factor and the terrain complexity and environmental complexity can be a preset mapping relationship. For example, the terrain complexity and environmental complexity are respectively input into the corresponding preset mapping sets to obtain the weight factors in the process of obtaining the data fusion requirement evaluation values corresponding to the terrain complexity and environmental complexity, and the mapping relationship therein can be one-to-one or many-to-one. In this embodiment, the value ranges of the terrain complexity weight factor and the environmental complexity weight factor are both limited to between 0 and 1, and the sum of the terrain complexity weight factor and the environmental complexity weight factor is 1.

[0052] In this embodiment, the data fusion requirement evaluation value is used to quantitatively evaluate the necessary degree of data fusion in the UAV surveying and mapping of the sub-region to be surveyed and mapped. The higher the terrain complexity or the environmental complexity, the greater the data fusion requirement evaluation value, indicating that the necessary degree of data fusion in the UAV surveying and mapping of the sub-region to be surveyed and mapped is higher.

[0053] The algorithm of this embodiment combines the terrain complexity and the environmental complexity, and comprehensively analyzes to obtain the evaluation value of the data fusion requirement. In this formula, the terrain complexity and the environmental complexity affect each other. As the terrain complexity increases, the environmental complexity will increase. For example, in mountainous areas, the complex terrain (such as high mountains and deep valleys) leads to large fluctuations in terrain, with high terrain complexity. Such terrain is prone to form unique microclimates. The air flow in the valleys is not smooth, humidity is easy to accumulate, and the deviation from the reference humidity outside increases. At the same time, the occlusion and reflection of light by the terrain are also more complex, which in turn leads to an increase in environmental complexity. As the environmental complexity increases, it will also conversely affect the degree of influence of terrain complexity on data collection. For example, in an environment with high humidity and strong electromagnetic interference, the performance of sensors used to measure the terrain will be affected, and the error in measuring the terrain complexity will increase, which means that the influence of the actual terrain complexity on surveying and mapping becomes more complex, and more accurate data fusion is required to compensate for the interference caused by environmental factors. By comprehensively analyzing the terrain complexity and the environmental complexity, the evaluation value of the data fusion requirement can be accurately obtained, which quantitatively reflects the actual data collection difficulty degree of the sub-region to be surveyed and mapped during the UAV surveying and mapping process due to the combined action of terrain and environmental factors.

[0054] Specifically, the number of data source fusions is judged according to the evaluation value of the data fusion requirement. The specific analysis process is as follows: The evaluation value of the data fusion requirement is matched with the preset data fusion requirement level intervals in the UAV surveying and mapping database to obtain the number of data source fusions corresponding to the preset data fusion requirement level intervals required for the sub-region to be surveyed and mapped; by subtracting the evaluation value of the data fusion requirement from the data fusion requirement threshold, the parameter adjustment determination value is obtained; the parameter adjustment determination value is compared with the preset sensor parameter adjustment threshold in the UAV surveying and mapping database: if the parameter adjustment determination value is greater than or equal to the preset sensor parameter adjustment threshold, the sensor parameters are adjusted; if the parameter adjustment determination value is less than the preset sensor parameter adjustment threshold, no additional operation is performed.

[0055] It should be understood that in this embodiment, multiple data fusion requirement level intervals are preset in the UAV surveying and mapping database, and each interval corresponds to a specific number of data source fusions. The evaluation value of the data fusion requirement of the sub-region to be surveyed and mapped is compared with each preset data fusion requirement level interval. When the evaluation value falls into a certain level interval, the number of data source fusions required for the corresponding sub-region to be surveyed and mapped can be obtained. For example, if the evaluation value of the data fusion requirement is within the level interval corresponding to the need to fuse 3 data sources, then it can be determined that the number of data source fusions required for the sub-region to be surveyed and mapped is 3.

[0056] Specifically, the sensor parameters are adjusted, and the specific analysis process is as follows: Obtain the preset terrain complexity threshold and environmental complexity threshold from the UAV mapping database. Subtract the terrain complexity from the terrain complexity threshold to obtain the terrain deviation; subtract the environmental complexity from the environmental complexity threshold to obtain the environmental deviation; match the specific adjustment amount through the terrain deviation, environmental deviation, and the preset parameter adjustment strategy table in the UAV mapping database, and adjust the sensitivity, resolution, and sampling frequency of the sensor.

[0057] In this embodiment, the greater the terrain deviation or environmental deviation, the greater the difference between the current terrain or environment and the preset standard, and it is necessary to adjust the sensor parameters to adapt to complex situations and improve the accuracy of data collection.

[0058] It should be understood that in this embodiment, for the sensitivity, if the terrain deviation is large, such as in mountainous areas or other terrain-complex regions, in order to obtain terrain details more accurately, the sensitivity of the sensor can be increased to make the sensor more sensitive to small terrain changes. If the environmental deviation is large, such as in areas with strong electromagnetic interference, the sensitivity of the sensor can be reduced to avoid excessive interference signals being collected, resulting in data distortion. For the resolution, when the terrain deviation is large, like in canyons or other places with drastic terrain undulations, the resolution of the sensor should be increased to obtain more detailed terrain data. When the environmental deviation is large, such as in areas with high humidity and poor lighting conditions, the resolution can be appropriately reduced because too high a resolution may collect a large amount of noise data, affecting the overall data quality. For the sampling frequency, in areas where the terrain or environment changes rapidly, that is, when the terrain deviation or environmental deviation is large, the sampling frequency needs to be increased to ensure that data changes can be captured in a timely manner. For example, in areas with frequent wind speed changes, increasing the sampling frequency can obtain more accurate wind speed data. In areas where the terrain and environment are relatively stable, the sampling frequency can be appropriately reduced to reduce data redundancy and processing burden. The specific increase or decrease is obtained by matching the terrain deviation, environmental deviation, and the preset parameter adjustment strategy table to obtain a specific adjustment amount.

[0059] It should be understood that in this embodiment, the terrain deviation, environmental deviation and the parameter adjustment strategy table preset in the UAV mapping database are matched to obtain specific adjustments to adjust the sensitivity, resolution and sampling frequency of the sensor. The parameter adjustment strategy table is constructed in the form of a two-dimensional matrix, the horizontal axis is a plurality of intervals of terrain deviation, and the vertical axis is a plurality of intervals of environmental deviation. Each cell inside the matrix stores the specific adjustments for different sensor parameters (sensitivity, resolution, sampling frequency) under the corresponding terrain and environmental deviation combination. For example, when the calculated terrain deviation falls into a certain interval of the horizontal axis, and the environmental deviation falls into a certain interval of the vertical axis, the sensor parameter adjustment amount suitable for the current terrain and environmental conditions can be found in the cell corresponding to the intersection of these two intervals. Taking sensitivity as an example, if the terrain deviation is large, such as in complex terrain areas such as mountainous areas, the adjustment amount for sensitivity is obtained by finding the cell corresponding to the terrain deviation interval and the current environmental deviation interval in the parameter adjustment strategy table, so as to improve the sensitivity of the sensor and make the sensor more sensitive to slight terrain changes. If the environmental deviation is large, such as in areas with strong electromagnetic interference, the corresponding adjustment amount is obtained from the parameter adjustment strategy table according to the above search method, so as to reduce the sensitivity of the sensor and avoid too many interference signals from being collected, resulting in data distortion. For resolution, when the terrain deviation is large, such as in places with dramatic terrain fluctuations such as canyons, the adjustment amount is matched from the parameter adjustment strategy table to improve the resolution of the sensor in order to obtain more detailed terrain data. When the environmental deviation is large, such as in areas with high humidity and poor lighting conditions, the resolution is appropriately reduced according to the adjustment amount of the corresponding position in the parameter adjustment strategy table, because too high a resolution may collect a large amount of noise data and affect the overall data quality. For sampling frequency, in areas where the terrain or environment changes rapidly, that is, when the terrain deviation or environmental deviation is large, the adjustment amount is found from the parameter adjustment strategy table to increase the sampling frequency to ensure that the data changes can be captured in time. For example, in areas where the wind speed changes frequently, increasing the sampling frequency can obtain more accurate wind speed data. In areas where the terrain and environment are relatively stable, the sampling frequency is appropriately reduced according to the adjustment amount given in the parameter adjustment strategy table to reduce data redundancy and processing burden.

[0060] The surveying and mapping data of the sub-area to be surveyed are fused according to the data fusion plan.

[0061] Specifically, the surveying and mapping data of the sub-region to be surveyed is subjected to data fusion according to the data fusion scheme. The specific analysis process is as follows: Obtain the UAV operation data, where the UAV operation data includes the UAV flight altitude, flight speed, and flight line overlap rate; Extract the preset UAV operation data reference values from the UAV surveying and mapping database, where the UAV operation data reference values include the reference flight altitude, allowable flight altitude deviation, reference flight speed, allowable flight speed deviation, reference flight line overlap rate, and allowable flight line overlap rate deviation; Perform a flight altitude allowable deviation correction operation on the flight altitude data of the sub-region to be fused and the preset reference flight altitude to obtain the flight altitude influence degree. The flight altitude allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree between the flight altitude data of the sub-region to be fused and the preset reference flight altitude corrected by the preset flight altitude weight factor; Perform a flight speed allowable deviation correction operation on the flight speed data of the sub-region to be fused and the preset reference flight speed to obtain the flight speed influence degree. The flight speed allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree between the flight speed data of the sub-region to be fused and the preset reference flight speed corrected by the preset flight speed weight factor; Perform a flight line overlap rate allowable deviation correction operation on the flight line overlap rate data of the sub-region to be fused and the preset reference flight line overlap rate to obtain the flight line overlap rate influence degree. The flight line overlap rate allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree between the flight line overlap rate data of the surveyed sub-region and the preset reference flight line overlap rate corrected by the preset flight line overlap rate weight factor; Obtain the operation status value by superimposing the flight altitude influence degree, flight speed influence degree, and flight line overlap rate influence degree; Perform a weighting process by combining the terrain complexity, environmental complexity, and operation status value to obtain the data source accuracy influence evaluation value of the sensor in the sub-region to be fused. The data source accuracy influence evaluation value represents the quantitative data of the influence degree of the terrain complexity, environmental complexity, flight altitude, flight speed, and flight line overlap rate on the quality of the data source of a certain sensor in the current sub-region to be fused; Sort according to the data source accuracy influence evaluation value to obtain the data source for data fusion.

[0062] It should be understood that in this embodiment, the flight altitude refers to the vertical distance of the UAV relative to the ground or a certain reference plane, which affects the surveying and mapping range and data resolution and can be measured by the Global Navigation Satellite System (GNSS). The flight speed refers to the speed of the UAV during flight, which affects the data acquisition efficiency and accuracy and can be measured by the Global Navigation Satellite System (GNSS). The flight line overlap rate refers to the overlapping degree between the flight lines of the UAV and can be obtained through the UAV flight planning software.

[0063] In a specific embodiment, the method for obtaining the data source accuracy influence evaluation value is as follows:

[0064] SJ tm = DX t *δ1 + HJ t *δ2 + FX t *δ3;

[0065]

[0066] δ1 + δ2 + δ3 = 1;

[0067] ε1 + ε2 = 1;

[0068] In the formula, SJ tm represents the evaluation value of the influence of the data source accuracy of the m-th sensor in the t-th sub-region to be fused, DX t represents the terrain complexity of the t-th sub-region to be fused, HJ t represents the environmental complexity of the t-th sub-region to be fused, FX t represents the operating status value of the t-th sub-region to be fused, FG t represents the flight altitude of the t-th sub-region to be fused, represents the preset reference flight altitude, ΔFG represents the preset allowable flight altitude deviation, FS t represents the flight speed of the t-th sub-region to be fused, represents the preset reference flight speed, ΔFS represents the preset allowable flight speed deviation, FC t represents the flight line overlap rate of the t-th sub-region to be fused, represents the preset reference flight line overlap rate, ΔFC represents the preset allowable flight line overlap rate deviation, δ1 represents the preset terrain complexity weight factor, δ2 represents the preset environmental complexity weight factor, δ3 represents the preset operating status value weight factor, ε1 represents the preset flight altitude weight factor, ε2 represents the preset flight speed weight factor, ε3 represents the preset flight line overlap rate weight factor, t represents the number of the sub-region to be fused, t = 1, 2, 3,..., g, g represents the total number of sub-regions to be fused, m represents the number of each sensor, m = 1, 2, 3,..., n, n represents the total number of sensors.

[0069] When δ1, δ2, and δ3 are used, the weight factors corresponding to the terrain complexity, environmental complexity, and operating status values in the process of obtaining the data source accuracy impact evaluation value can be directly obtained from the UAV mapping database, and they respectively represent the numerical values of the influence degrees of the terrain complexity, environmental complexity, and operating status values on the data source accuracy impact evaluation value. The terrain complexity weight factor, environmental complexity weight factor, and operating status value weight factor and the terrain complexity, environmental complexity, and operating status value can be a preset mapping relationship. For example, by inputting the terrain complexity, environmental complexity, and operating status value into the corresponding preset mapping sets respectively, the weight factors in the process of obtaining the data source accuracy impact evaluation value corresponding to the terrain complexity, environmental complexity, and operating status value are obtained, and the mapping relationship therein can be one-to-one or many-to-one. In this embodiment, the value ranges of the terrain complexity weight factor, environmental complexity weight factor, and operating status value weight factor are all limited between 0 and 1, and the sum of the terrain complexity weight factor, environmental complexity weight factor, and operating status value weight factor is 1.

[0070] When ε1, ε2, and ε3 are used, the weight factors corresponding to the flight altitude, flight speed, and flight line overlap rate in the process of obtaining the operating status value can be directly obtained from the UAV mapping database, and they respectively represent the numerical values of the influence degrees of the flight altitude, flight speed, and flight line overlap rate on the operating status value. The flight altitude weight factor, flight speed weight factor, and flight line overlap rate weight factor and the flight altitude, flight speed, and flight line overlap rate can be a preset mapping relationship. For example, by inputting the flight altitude, flight speed, and flight line overlap rate into the corresponding preset mapping sets respectively, the weight factors corresponding to the flight altitude, flight speed, and flight line overlap rate in the process of obtaining the operating status value are obtained, and the mapping relationship therein can be one-to-one or many-to-one. In this embodiment, the value ranges of the flight altitude weight factor, flight speed weight factor, and flight line overlap rate weight factor are all limited between 0 and 1, and the sum of the flight altitude weight factor, flight speed weight factor, and flight line overlap rate weight factor is 1.

[0071] In this embodiment, the data source accuracy impact evaluation value is used to quantitatively evaluate the quality impact degree of the data source of a certain sensor in the current sub-region to be fused. The higher the terrain complexity, or the higher the environmental complexity, or the larger the operating status value, the larger the data source accuracy impact evaluation value, indicating that the quality impact degree of the data source of the sensor in the current sub-region to be fused is higher.

[0072] In this embodiment, the operation status value is used to quantitatively evaluate the influence degree of the flight status of the UAV in the sub-region to be fused on the quality of the sensor data source. The greater the deviation of the flight altitude from the reference flight altitude, or the greater the deviation of the flight speed from the reference flight speed, or the greater the deviation of the flight line overlap rate from the reference flight line overlap rate, the greater the operation status value, indicating that the influence degree of the flight status of the UAV in the sub-region to be fused on the quality of the sensor data source is higher.

[0073] The algorithm of this embodiment combines the terrain complexity, environmental complexity and operation status value, and comprehensively analyzes to obtain the evaluation value of the influence on the data source accuracy. In this formula, the terrain complexity, environmental complexity and operation status value affect each other. As the operation status value increases, in complex terrains and environments, the influence on the data source accuracy is more significant. For example, when the UAV has a higher flight altitude or faster speed, in a complex terrain area, the sensor may not be able to accurately obtain terrain detail information, and in a complex environment, the interference of environmental factors on the data will be amplified due to the unstable operation status, resulting in an increase in the evaluation value of the influence on the data source accuracy. By comprehensively analyzing the terrain complexity, environmental complexity and operation status value, the evaluation value of the influence on the data source accuracy can be accurately obtained, which quantitatively reflects the deviation degree between the quality of the sensor data source and the quality of the data source in the ideal state during the UAV mapping process due to the combined action of terrain, environment and flight status.

[0074] The algorithm of this embodiment combines the flight altitude, flight speed and flight line overlap rate, and comprehensively analyzes to obtain the operation status value. In this formula, the flight altitude, flight speed and flight line overlap rate affect each other. As the deviation of the flight altitude from the reference flight altitude increases, it will affect the deviation of the flight speed from the reference flight speed, and at the same time affect the deviation of the flight line overlap rate from the reference flight line overlap rate. As the deviation of the flight speed from the reference flight speed increases, the deviation of the flight altitude from the reference flight altitude and the deviation of the flight line overlap rate from the reference flight line overlap rate will also be affected. If the flight speed increases, in order to avoid too sparse data collection, the flight altitude may need to be reduced to narrow the sensor scanning range. At the same time, due to the increase in flight speed, the area passed by the UAV per unit time increases. To ensure that no data is missed, the flight line overlap rate needs to be increased. For example, when conducting forest fire monitoring, in order to quickly obtain the fire spread situation, the UAV increases its flight speed, correspondingly reduces its flight altitude, and increases the flight line overlap rate. By comprehensively analyzing the flight altitude, flight speed and flight line overlap rate, the operation status value can be accurately obtained, which quantitatively reflects the deviation degree between the actual status and the ideal reference status of the flight altitude, flight speed and flight line overlap rate when the UAV is flying in the sub-region to be fused, as well as the comprehensive influence degree on the quality of the sensor data source.

[0075] Specifically, sort according to the data source accuracy impact evaluation value, obtain the data sources for data fusion, and the specific analysis process is as follows: Sort the data source accuracy impact evaluation values of each sensor from low to high; Compare the data source accuracy impact evaluation value with the preset data source accuracy impact evaluation threshold in the UAV mapping database, count the number of data sources less than or equal to the data source accuracy impact evaluation threshold, and mark it as the number of candidate data sources for fusion; Compare the number of candidate data sources for fusion with the number of data sources for fusion. If the number of candidate data sources for fusion is greater than or equal to the number of data sources for fusion, select the data sources for data fusion in ascending order according to the data source accuracy impact evaluation value; If the number of candidate data sources for fusion is less than the number of data sources for fusion, adjust the UAV parameters and sensor parameters for secondary acquisition; If the acquisition time is insufficient, reduce the data source accuracy requirement and increase the number of data sources for fusion.

[0076] It should be understood that in this embodiment, each sensor will generate a data source accuracy impact evaluation value, which reflects the impact degree of the data source of this sensor on the final data accuracy. Arrange the data source accuracy impact evaluation values of all sensors in ascending order, compare each data source accuracy impact evaluation value with the data source accuracy impact evaluation threshold, count the number of data sources less than or equal to the data source accuracy impact evaluation threshold, and mark it as the number of candidate data sources for fusion. Compare the counted number of candidate data sources for fusion with the number of data sources for fusion required for the sub-region to be mapped determined previously. If the number of candidate data sources for fusion is greater than or equal to the required number of data sources for fusion, at this time, select the data sources with lower data source accuracy impact evaluation values for data fusion according to the previous ascending order, and the number of selected data sources is the number of data sources for fusion required for the sub-region to be mapped determined previously. If the number of candidate data sources for fusion is less than the required number of data sources for fusion, it means that the number of existing data sources with qualified accuracy is insufficient. In this case, it is necessary to adjust the UAV parameters and sensor parameters, and then perform secondary data acquisition. If it is found that the acquisition time is insufficient during the secondary acquisition and it is impossible to obtain data sources that meet the accuracy requirements according to the original plan. At this time, in order to complete the data fusion, reduce the requirements for the data source accuracy and increase the number of data sources for fusion to ensure the smooth progress of the data fusion.

[0077] Specifically, the UAV parameters and sensor parameters are adjusted. The specific analysis process is as follows: By subtracting the number of candidate data sources that can be fused from the number of data source fusions, the data source supply-demand difference is obtained; the data source supply-demand difference is compared with the preset difference adjustment threshold interval in the UAV mapping database. If the data source supply-demand difference is within the low difference adjustment threshold interval, only the sensor parameters are adjusted; if the data source supply-demand difference is within the high difference adjustment threshold interval, the UAV parameters and sensor parameters are adjusted synchronously; according to the data source supply-demand difference and the preset sensor parameter adjustment amplitude and UAV parameter adjustment amplitude in the UAV mapping database, the specific adjustment amount is matched, and the sensitivity, resolution, and sampling frequency of the sensor and the flight altitude, flight speed, and flight line overlap rate of the UAV are adjusted.

[0078] It should be understood that in this embodiment, the low difference adjustment threshold interval and the high difference adjustment threshold interval are the difference adjustment threshold intervals preset in the UAV mapping database. For example, assume that the preset low difference adjustment threshold interval is [-3, 3]. When the calculated data source supply-demand difference is within this range, it indicates that the current data source supply situation is relatively good. By only adjusting the sensor parameters, such as appropriately increasing the sensitivity, resolution, or sampling frequency of the sensor, it may be possible to meet the requirements for the quantity and quality of data sources in data fusion without adjusting the parameters of the UAV, which can reduce the impact on the UAV flight mission and lower the adjustment cost and complexity; the preset high difference adjustment threshold interval is (-∞, -3) ∪ (3, +∞). When the data source supply-demand difference is greater than 3 or less than -3, it indicates that relying solely on adjusting the sensor parameters may not effectively solve the problem of insufficient data sources or substandard quality, and it is necessary to adjust the parameters of the UAV simultaneously, such as changing the flight altitude, flight speed, and flight line overlap rate of the UAV.

[0079] It should be understood that in this embodiment, based on the comparison between the data source supply-demand difference and the preset threshold range, it is determined whether to only adjust the sensor parameters or simultaneously adjust the UAV and sensor parameters to meet the requirement of the number of data source fusions and ensure the quality of the surveying and mapping data. If the difference is within the low-difference adjustment threshold range, it indicates that the number of data sources is only slightly insufficient, so only the sensor parameters need to be adjusted to obtain more data; if the difference is within the high-difference adjustment threshold range, it indicates that the number of data sources is seriously insufficient, and it is necessary to simultaneously adjust the UAV parameters and sensor parameters to improve the data acquisition efficiency. According to the matching between the data source supply-demand difference and the preset adjustment amplitudes of the sensor parameters and UAV parameters, the specific adjustment amount is determined. For sensitivity, when it is necessary to enhance the sensor's ability to capture the target signal, the sensitivity is increased. For example, in an environment with insufficient light or weak signals, the camera ISO or radar signal gain is increased to enhance the signal strength. For resolution, to obtain detailed data, the resolution is increased. For example, the camera pixel density or lidar point cloud density is increased. When surveying areas with high requirements for terrain details, the resolution can be appropriately increased. For the sampling frequency, to ensure continuous and complete data acquisition in a dynamic environment or complex terrain, the sampling frequency is increased. For example, when surveying an area with large terrain undulations, the camera shooting frequency or lidar scanning frequency is increased. For the flight altitude, reducing the flight altitude can improve the data resolution and accuracy because the sensor can obtain detailed information closer to the ground. For example, when surveying a small area with high accuracy requirements, the flight altitude can be reduced. For the flight speed, reducing the flight speed helps the sensor collect data more stably, especially in cases where high-precision data is required, such as in complex terrains or areas with extremely high requirements for data accuracy, the flight speed is reduced. For the flight line overlap rate, increasing the flight line overlap rate can improve the data redundancy and integrity, which is beneficial for subsequent data fusion and processing. For example, when performing tasks such as terrain surveying and building modeling, the flight line overlap rate is increased.

[0080] It should be understood that in this embodiment, the specific adjustment amount is determined by matching the data source supply-demand difference with the preset adjustment ranges of sensor parameters and UAV parameters. For sensor parameters and UAV parameters in the UAV mapping database, multiple adjustment range intervals corresponding to the data source supply-demand difference are preset respectively. After obtaining the data source supply-demand difference, first determine the range where the difference is located. For example, for sensor sensitivity, if the data source supply-demand difference falls within a certain range, and the preset sensitivity adjustment range corresponding to this range is "increase by 10%-20%", then according to the specific position of the difference within this range, determine the specific increase ratio through methods such as linear interpolation. Suppose the data source supply-demand difference is at the middle position of this range, and after calculation, the sensor sensitivity is increased by 15%. For resolution, in the same way, if the data source supply-demand difference corresponds to a preset resolution adjustment range of "increase the camera pixel density by 5%-10%", determine the specific increase value according to the relative position of the difference within this range. For example, if the difference is close to the upper limit of the range, the camera pixel density is increased by about 8%. For the sampling frequency, if the preset adjustment range corresponding to the data source supply-demand difference is "increase the lidar scanning frequency by 1-3 times per second", determine the specific increase times according to the position of the difference within this interval. For the flight altitude of the UAV, if the preset adjustment range corresponding to the data source supply-demand difference is "decrease by 5-10 meters", determine the decreased altitude according to the position of the difference. For example, if the difference is at the rear, it is decreased by 8 meters. For the flight speed, if the preset adjustment range is "decrease by 5-10 km / h", determine the specific decreased speed value according to the position of the difference within this range. If the difference is in the middle, it is decreased by 7 km / h. For the flight line overlap rate, if the preset adjustment range is "increase by 5%-10%", determine the specific increase ratio according to the position of the difference within the corresponding interval. If the difference is towards the front end, it is increased by 6%. Through the above methods, accurately match the data source supply-demand difference with the preset adjustment range, and determine the specific adjustment amounts of various parameters of the sensor and the UAV, so as to achieve more efficient and accurate data collection, meet the requirements of the number of data source fusions, and ensure the quality of mapping data.

[0081] Specifically, to reduce the data source accuracy requirements and increase the number of data source fusions, the specific analysis process is as follows: Calculate the average value of the data source accuracy impact evaluation for the sub-region to be fused based on the data source accuracy impact evaluation values of all sensors in the sub-region to be fused. Subtract the preset data source accuracy impact evaluation threshold from the average value of the data source accuracy impact evaluation, and correct the data source accuracy impact evaluation threshold according to the difference. Compare the data source accuracy impact evaluation value with the corrected data source accuracy impact evaluation threshold, obtain the data sources that are less than or equal to the corrected data source accuracy impact evaluation threshold, and start a high-level fusion algorithm for data fusion.

[0082] It should be understood that, in this embodiment, the data source accuracy impact assessment values ​​of all sensors in the sub-area to be fused are calculated to obtain the data source accuracy impact assessment mean of the sub-area to be fused, and the difference obtained by subtracting the data source accuracy impact assessment mean from the preset data source accuracy impact assessment threshold is used to correct the preset data source accuracy impact assessment threshold, and the absolute value of the difference is superimposed on the data source accuracy impact assessment threshold to obtain the corrected data source accuracy impact assessment threshold, and the data source accuracy impact assessment value is compared with the corrected threshold again to screen out all data sources that are less than or equal to the corrected data source accuracy impact assessment threshold.

[0083] In the drone mapping of this embodiment, different fusion algorithms have different data processing capabilities and precisions. Low-level fusion algorithms may only perform simple data superposition or preliminary feature extraction, which is suitable for situations where the data quality is good and the environment is relatively simple. High-level fusion algorithms usually use more complex mathematical models and deep learning techniques, such as fusion algorithms based on neural networks. Starting a high-level fusion algorithm means that when the data source accuracy is low or the data fusion requirements are complex, these advanced algorithms are used to more effectively integrate data from multiple sensors, which can better handle the correlation and complementarity between data, remove noise and redundant information, thereby improving the accuracy and reliability of data fusion, and obtaining more accurate and more realistic mapping results. For example, in complex terrain and harsh environments, the data collected by multiple sensors may have errors and interference. High-level fusion algorithms can deeply process these data, mine the potential information in the data, and provide more powerful support for subsequent mapping analysis and decision-making.

[0084] The UAV mapping database is used to store various data related to UAV mapping, including critical elevation, critical slope, reference light intensity, allowable light intensity deviation, critical wind speed, reference flight altitude, allowable flight altitude deviation, parameter adjustment strategy table, and sensor parameter adjustment range. The data in the UAV mapping database can be obtained from terrain sensors, elevation meters, meteorological station equipment, electromagnetic interference monitors, UAV flight parameter recording modules, monitoring systems of various mapping sensors, and professional UAV mapping data processing and analysis software.

[0085] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. As long as it does not deviate from the structure of the present invention or exceed the scope defined by the present invention, it should fall within the protection scope of the present invention.

Claims

1. A data fusion and optimization method in UAV mapping based on deep learning, characterized in that: Including: Dividing the surveying and mapping area to obtain sub-areas to be surveyed, and collecting the surveying and mapping complexity data of the sub-areas to be surveyed. The surveying and mapping complexity data includes environmental data and geographical data; Processing the surveying and mapping complexity data of the sub-areas to be surveyed to obtain a data fusion requirement determination result, and determining a data fusion scheme based on the data fusion requirement determination result. The data fusion scheme is used to determine the number of data sources for data fusion; Performing data fusion on the surveying and mapping data of the sub-areas to be surveyed according to the data fusion scheme.

2. The data fusion and optimization method in drone mapping based on deep learning according to claim 1, wherein: The geographical data includes elevation, slope, terrain undulation degree, and surface roughness, and the environmental data includes light intensity, humidity, wind speed, visibility, and electromagnetic interference intensity; The process of processing the surveying and mapping complexity data of the sub-areas to be surveyed to obtain a data fusion requirement determination result is as follows: Extracting preset surveying and mapping complexity data reference values from the UAV surveying and mapping database. The surveying and mapping complexity data reference values include critical elevation, critical slope, critical terrain undulation degree, critical surface roughness, reference light intensity, allowable light intensity deviation value, reference humidity, allowable humidity deviation value, critical wind speed, critical visibility, and critical electromagnetic interference intensity; Processing the geographical data and environmental data respectively to obtain the terrain complexity and environmental complexity of the sub-areas to be surveyed; Processing the terrain complexity and environmental complexity to obtain the data fusion requirement evaluation value of the sub-areas to be surveyed; Comparing the data fusion requirement evaluation value with the preset data fusion requirement evaluation threshold in the UAV surveying and mapping database: If the data fusion requirement evaluation value is less than the preset data fusion requirement evaluation threshold, the data fusion requirement determination result of the sub-areas to be surveyed is recorded as not performing data fusion; If the data fusion requirement evaluation value is greater than or equal to the preset data fusion requirement evaluation threshold, the data fusion requirement determination result of the sub-areas to be surveyed is recorded as performing data fusion, and it is marked as a sub-area to be fused. The number of data sources for fusion is judged according to the data fusion requirement evaluation value.

3. The data fusion and optimization method in drone mapping based on deep learning according to claim 2, characterized in that: The process of processing the geographical data and environmental data respectively to obtain the terrain complexity and environmental complexity of the sub-areas to be surveyed is as follows: Performing an elevation ratio approaching influence degree operation on the elevation data of the sub-areas to be surveyed and the preset critical elevation to obtain an elevation influence degree. The elevation ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the elevation data of the sub-areas to be surveyed and the preset critical elevation corrected by the preset elevation weight factor; Performing a slope ratio approaching influence degree operation on the slope data of the sub-areas to be surveyed and the preset critical slope to obtain a slope influence degree. The slope ratio approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the slope data of the sub-areas to be surveyed and the preset critical slope corrected by the preset slope weight factor; The terrain undulation influence degree is obtained by performing a terrain undulation proportion approaching influence degree operation on the terrain undulation data of the sub-region to be surveyed and the preset critical terrain undulation. The terrain undulation proportion approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the terrain undulation data of the sub-region to be surveyed and the preset critical terrain undulation corrected by the preset terrain undulation weight factor; The surface roughness influence degree is obtained by performing a surface roughness proportion approaching influence degree operation on the surface roughness data of the sub-region to be surveyed and the preset critical surface roughness. The surface roughness proportion approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the surface roughness data of the sub-region to be surveyed and the preset critical surface roughness corrected by the preset surface roughness weight factor; The terrain complexity of the sub-region to be surveyed is obtained by superimposing the elevation influence degree, slope influence degree, terrain undulation influence degree and surface roughness influence degree. The terrain complexity represents the quantitative data of the complexity of the elevation, slope, terrain undulation and surface roughness for the geographical factors on the sub-region to be surveyed; The light influence degree is obtained by performing a light intensity allowable deviation correction operation on the light intensity data of the sub-region to be surveyed and the preset reference light intensity. The light intensity allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree of the light intensity data of the sub-region to be surveyed and the preset reference light intensity corrected by the preset light weight factor; The humidity influence degree is obtained by performing a humidity allowable deviation correction operation on the humidity data of the sub-region to be surveyed and the preset reference humidity. The humidity allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree of the humidity data of the sub-region to be surveyed and the preset reference humidity corrected by the preset humidity weight factor; The wind speed influence degree is obtained by performing a wind speed proportion approaching influence degree operation on the wind speed data of the sub-region to be surveyed and the preset critical wind speed. The wind speed proportion approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the wind speed data of the sub-region to be surveyed and the preset critical wind speed corrected by the preset wind speed weight factor; The visibility influence degree is obtained by performing a visibility proportion approaching influence degree operation on the visibility data of the sub-region to be surveyed and the preset critical visibility. The visibility proportion approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the visibility data of the sub-region to be surveyed and the preset critical visibility corrected by the preset visibility weight factor; The electromagnetic interference influence degree is obtained by performing an electromagnetic interference intensity proportion approaching influence degree operation on the electromagnetic interference intensity data of the sub-region to be surveyed and the preset critical electromagnetic interference intensity. The electromagnetic interference intensity proportion approaching influence degree operation represents the quantitative data of the complexity influence corresponding to the approaching degree of the electromagnetic interference intensity data of the sub-region to be surveyed and the preset critical electromagnetic interference intensity corrected by the preset electromagnetic interference weight factor; The environmental complexity of the sub-region to be surveyed and mapped is obtained by superimposing the illumination influence degree, humidity influence degree, wind speed influence degree, visibility influence degree, and electromagnetic interference influence degree. The environmental complexity represents the quantitative data of the complexity degree of illumination intensity, humidity, wind speed, visibility, and electromagnetic interference intensity on the environmental factors for the sub-region to be surveyed and mapped.

4. The data fusion and optimization method in drone mapping based on deep learning according to claim 2, characterized in that: The evaluation value of the data fusion requirement for the sub-region to be surveyed and mapped is obtained by processing the terrain complexity and environmental complexity. The specific analysis process is as follows: The terrain influence degree is obtained by weighting the terrain complexity, the environmental influence degree is obtained by weighting the environmental complexity, and the evaluation value of the data fusion requirement for the sub-region to be surveyed and mapped is obtained by superimposing the terrain influence degree and the environmental influence degree; The evaluation value of the data fusion requirement represents the quantitative data of the necessary degree of terrain complexity and environmental complexity for data fusion in UAV surveying and mapping of the sub-region to be surveyed and mapped.

5. The data fusion and optimization method in drone mapping based on deep learning according to claim 4, wherein: The number of data source fusions is determined according to the evaluation value of the data fusion requirement. The specific analysis process is as follows: Match the evaluation value of the data fusion requirement with the preset data fusion requirement level interval in the UAV surveying and mapping database to obtain the number of data source fusions corresponding to the preset data fusion requirement level interval required for the sub-region to be surveyed and mapped; By subtracting the evaluation value of the data fusion requirement from the data fusion requirement evaluation threshold, the parameter adjustment determination value is obtained; Compare the parameter adjustment determination value with the preset sensor parameter adjustment threshold in the UAV surveying and mapping database: If the parameter adjustment determination value is greater than or equal to the preset sensor parameter adjustment threshold, adjust the sensor parameters; If the parameter adjustment determination value is less than the preset sensor parameter adjustment threshold, no additional operation is performed.

6. The data fusion and optimization method in UAV mapping based on deep learning according to claim 5, characterized in that: The adjustment of the sensor parameters, the specific analysis process is as follows: Obtain the preset terrain complexity threshold and environmental complexity threshold from the UAV surveying and mapping database. By subtracting the terrain complexity from the terrain complexity threshold, the terrain deviation degree is obtained; By subtracting the environmental complexity from the environmental complexity threshold, the environmental deviation degree is obtained; Match the specific adjustment amount through the terrain deviation degree, environmental deviation degree and the preset parameter adjustment strategy table in the UAV surveying and mapping database to adjust the sensitivity, resolution and sampling frequency of the sensor.

7. The data fusion and optimization method in drone mapping based on deep learning according to claim 1, characterized in that: The surveying and mapping data of the sub-region to be surveyed and mapped are fused according to the data fusion scheme. The specific analysis process is as follows: Obtain the UAV operation data, and the UAV operation data includes the UAV flight height, flight speed, and flight line overlap rate; Extract the preset UAV operation data reference values from the UAV surveying and mapping database. The UAV operation data reference values include the reference flight height, allowable flight height deviation, reference flight speed, allowable flight speed deviation, reference flight line overlap rate, and allowable flight line overlap rate deviation; The flight height influence degree is obtained by performing a flight height allowable deviation correction operation on the flight height data of the sub-region to be fused and the preset reference flight height. The flight height allowable deviation correction operation represents the quantitative data of the complexity influence corresponding to the deviation degree between the flight height data of the sub-region to be fused and the preset reference flight height corrected by the preset flight height weight factor; The flight speed influence degree is obtained by performing a flight speed allowable deviation correction operation on the flight speed data of the sub-region to be fused and a preset reference flight speed. The flight speed allowable deviation correction operation represents the quantization data of the complexity influence corresponding to the deviation degree between the flight speed data of the sub-region to be fused and the preset reference flight speed, which is corrected by a preset flight speed weight factor. The flight path overlap rate influence degree is obtained by performing a flight path overlap rate allowable deviation correction operation on the flight path overlap rate data of the sub-region to be fused and a preset reference flight path overlap rate. The flight path overlap rate allowable deviation correction operation represents the quantization data of the complexity influence corresponding to the deviation degree between the flight path overlap rate data of the sub-region to be surveyed and mapped and the preset reference flight path overlap rate, which is corrected by a preset flight path overlap rate weight factor. The operation state value is obtained by superimposing the flight altitude influence degree, the flight speed influence degree, and the flight path overlap rate influence degree. By combining the terrain complexity, the environmental complexity, and the operation state value for weight assignment processing, the data source accuracy influence evaluation value of the sensor in the sub-region to be fused is obtained. The data source accuracy influence evaluation value represents the quantization data of the influence degree of the terrain complexity, the environmental complexity, the flight altitude, the flight speed, and the flight path overlap rate on the quality of the data source of a certain sensor in the current sub-region to be fused. Sort according to the data source accuracy influence evaluation value to obtain the data sources for data fusion.

8. The data fusion and optimization method in UAV mapping based on deep learning according to claim 7, characterized in that: The specific analysis process of sorting according to the data source accuracy influence evaluation value to obtain the data sources for data fusion is as follows: Sort the data source accuracy influence evaluation values of each sensor from low to high. Compare the data source accuracy influence evaluation value with a preset data source accuracy influence evaluation threshold in the UAV mapping database, count the number of data sources less than or equal to the data source accuracy influence evaluation threshold, and mark it as the number of candidate data sources for fusion. Compare the number of candidate data sources for fusion with the number of data sources for fusion. If the number of candidate data sources for fusion is greater than or equal to the number of data sources for fusion, select the data sources for data fusion according to the data source accuracy influence evaluation value sorted from low to high. If the number of candidate data sources for fusion is less than the number of data sources for fusion, adjust the UAV parameters and sensor parameters for secondary acquisition. If the acquisition time is insufficient, reduce the data source accuracy requirement and increase the number of data sources for fusion.

9. The data fusion and optimization method in drone mapping based on deep learning according to claim 8, characterized in that: The specific analysis process of adjusting the UAV parameters and sensor parameters is as follows: Obtain the data source supply-demand difference by subtracting the number of candidate data sources for fusion from the number of data sources for fusion. Compare the data source supply-demand difference with a preset difference adjustment threshold range in the UAV mapping database. If the data source supply-demand difference is in the low difference adjustment threshold range, only adjust the sensor parameters. If the data source supply-demand difference is in the high difference adjustment threshold range, synchronously adjust the UAV parameters and sensor parameters. Match the specific adjustment amount according to the data source supply-demand difference and the preset sensor parameter adjustment range and UAV parameter adjustment range in the UAV mapping database, and adjust the sensitivity, resolution, and sampling frequency of the sensor and the flight altitude, flight speed, and flight path overlap rate of the UAV.

10. The data fusion and optimization method in drone mapping based on deep learning according to claim 8, characterized in that: The requirements for the accuracy of the data source are reduced, and the number of data source fusions is increased. The specific analysis process is as follows: Based on the evaluation values of the data source accuracy impacts of all sensors in the sub-region to be fused, the average evaluation value of the data source accuracy impacts of the sub-region to be fused is calculated. By subtracting the preset evaluation threshold of the data source accuracy impact from the average evaluation value of the data source accuracy impact, the evaluation threshold of the data source accuracy impact is corrected according to the difference value; Compare the evaluation value of the data source accuracy impact with the corrected evaluation threshold of the data source accuracy impact, obtain the data sources that are less than or equal to the corrected evaluation threshold of the data source accuracy impact, and start a high-level fusion algorithm for data fusion.

Images