Unmanned aerial vehicle route big data planning method and system for water conservancy surveying and mapping

Abstract

The invention relates to the technical field of route planning, and discloses an unmanned aerial vehicle route big data planning method and system for water conservancy surveying and mapping, and the method comprises the steps: S1, inputting a route starting position, a route ending position and a water conservancy surveying and mapping terrain range, and obtaining a plurality of unmanned aerial vehicle flight routes passing through a water conservancy surveying and mapping region after analysis; s2, determining a plurality of estimated consumed paths of the flight routes of the unmanned aerial vehicle passing through the water conservancy surveying and mapping area and energy consumption parameters under the estimated consumed paths, and then giving a unique flight route according to matching of maximum energy consumption capable of being provided by the battery capacity of the unmanned aerial vehicle and the energy consumption parameters; s3, collecting illumination intensity data and spectral distribution data of sunlight in real time to obtain original data; and S4, preprocessing the original data to obtain a data set. According to the system, by integrating a plurality of functional modules, intelligent planning and management of the route of the unmanned aerial vehicle are realized, and powerful technical support is provided for water conservancy surveying and mapping work.

Description

Technical Field

[0001] The present invention relates to the technical field of route planning, and in particular to a method and system for planning UAV route big data for water conservancy surveying and mapping. Background Art

[0002] Drones are widely used in water conservancy surveying and mapping because they have the characteristics of fast flight speed and wide coverage area. Through large-scale flight and rapid inspection, water resources survey information can be obtained in the first place, and water conservancy surveying and mapping data can be obtained.

[0003] For example, Chinese patent publication number: CN112249323A discloses a water conservancy project drone for remote sensing mapping; Another example is Chinese patent publication number: CN118358788A, which discloses a water conservancy surveying and mapping drone that adjusts to the environment.

[0004] In the prior art including the above patents, when planning the route of a drone during cruising, the route with the shortest time or distance is often calculated based on the waypoints, but it lacks consideration of the drone's endurance and cannot obtain a route plan that is more conducive to terrain mapping. To this end, we propose a drone route big data planning method and system for water conservancy surveying and mapping. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for planning UAV route big data for water conservancy surveying and mapping, so as to overcome the technical problems existing in the prior art.

[0006] In order to achieve the above technical objectives and the above technical effects, the present invention provides the following technical solutions: A method for planning UAV route big data for water conservancy surveying and mapping, comprising the following steps: S1. Input the starting position of the route, the ending position of the route and the water conservancy surveying and mapping terrain range, and obtain multiple UAV flight routes passing through the water conservancy surveying and mapping area after analysis; S2, determining the estimated consumption distances of multiple UAV flight routes passing through the water conservancy surveying and mapping area and the energy consumption parameters under the estimated consumption distances, and then matching the maximum energy consumption that can be provided by the UAV battery capacity with the energy consumption parameters to give a unique flight route; S3, collecting sunlight intensity data and spectral distribution data in real time to obtain original data; S4, performing preprocessing on the original data to obtain a data set; S5. Input the data set into the prediction model to obtain the total photovoltaic power generation in the future predetermined period; S6, adjusting the maximum energy consumption based on the total amount of photovoltaic power generation in the future predetermined period, and judging the energy consumption parameters under the only flight route, generating an execution strategy based on the judgment result, and uploading it to the drone main control module to generate a corresponding execution command; S7, based on the unique flight route given in step S02, the terrain below the route is monitored and stored in the camera unit during the flight; S8, extracting the camera data by using the drawing unit and drawing the regional terrain at a reduced scale; S9. Upload the route data and surveying model to the cloud database.

[0007] Preferably, the preprocessing of the raw data in step S4 includes: S41, obtaining a sequence of light intensity data and spectral distribution data of the original data; S42, based on the IQR method, respectively detect and mark the abnormal values ​​of the sequence of the light intensity data and the spectral distribution data, wherein the abnormal values ​​are values ​​lower than Q1-1.5*IQR and Q3+1.5*IQR, wherein: Q1=the 25th percentile value of the sequence of the light intensity data, Q3=the 75th percentile value of the sequence of the spectral distribution data; S43, for the identified abnormal value, select the average value of the first three adjacent data in sequence to replace it, so as to obtain the cleaned light intensity data and the cleaned spectral distribution data.

[0008] Preferably, the preprocessing of the cleaned light intensity data and the cleaned spectral distribution data in step S4 further includes: S44, feature extraction of light intensity data after cleaning: S441, calculating the maximum light intensity value in each predetermined time period to obtain a peak value; S442, calculating the average value of multiple light intensities in each predetermined time period; S443, calculating the difference between the maximum value and the minimum value of the light intensity in each predetermined time period to obtain the maximum difference of the light intensity in each predetermined time period; S44. Feature extraction of spectral distribution data after cleaning: S444, selecting a specific wavelength as the light intensity value to obtain the light intensity value at the key wavelength; S445, calculating the integral of the illumination intensity within the entire spectral range to obtain a spectral integral value; S446. Find the wavelength corresponding to the maximum light intensity in the spectral distribution to obtain the peak wavelength value.

[0009] Preferably, the prediction model in step S5 is generated by defining a prediction model architecture comprising an input layer, a one-dimensional convolution layer, an activation function layer, a pooling layer, a flattening layer, a fully connected layer and an output layer, wherein: The shape of the input layer matches the dimensions of the spectral data; The one-dimensional convolution layer uses the preset convolution kernel size and number to extract spectral features; The activation function layer uses the ReLU function to increase nonlinearity; The pooling layer uses maximum pooling or average pooling to reduce the feature dimension; The flattening layer flattens the multi-dimensional feature map into a one-dimensional vector; The fully connected layer sets the appropriate number of layers and neurons according to the extracted and cleaned light intensity data features and the extracted and cleaned spectral distribution data features; The output layer is a linear regressor.

[0010] Preferably, in step S5, inputting the data set into the prediction model for training includes: S51, inputting the extracted light intensity data features after cleaning and the extracted spectral distribution data features after cleaning into the prediction model, performing forward calculation through the input layer, the one-dimensional convolution layer, the activation function layer, the pooling layer, and the fully connected layer to obtain a predicted output value; S52, calculating the loss value of the predicted output value and the true label based on the mean square error; S53, calculating the gradient of each layer according to the loss value, and transferring the gradient from the output layer to the input layer through the back propagation algorithm; S54, using the optimizer and the calculated gradient to update the weight of the model to obtain a new prediction model, and the predicted output value is used as the true label of the next set of data; wherein the purpose of the weight update is to reduce the loss value; S55. Obtain a forecast data set based on adjacent differences between at least three sets of data as the growth rate, and input it into the new forecast model to obtain the total photovoltaic power generation within a future predetermined period.

[0011] Preferably, in step S6, adjusting the maximum energy consumption based on the total amount of photovoltaic power generation obtained in the future predetermined period, and judging the energy consumption parameters under the only flight route include: S61, collecting real-time energy consumption of the UAV during operation; S62, collecting the remaining total amount of the current drone battery; S63, based on the obtained total photovoltaic power generation in the future predetermined period and the energy consumption, when the difference is a negative value, the remaining total amount of the current drone battery is used to compensate to zero; S64: When the remaining total amount of the current drone battery is equal to the energy consumption of the drone after completing the unique flight route, the result of the determination is return flight.

[0012] A UAV route big data planning system for water conservancy surveying and mapping, which is used to implement a UAV route big data planning method for water conservancy surveying and mapping described in the above scheme, comprising: A route acquisition module, which is used to set the starting point and the terminal of the target mapping area, generate multiple flight routes within the target mapping area, and provide estimated energy consumption parameters under a unique flight route; Energy storage consumption statistics module, real-time acquisition of actual flight consumption under a unique flight route; Solar power supply module and prediction module, which predicts future photovoltaic energy replenishment data based on the collected light intensity data and spectral distribution data; A data processing module, which obtains the adjusted estimated energy consumption parameters based on the actual flight consumption and the predicted future photovoltaic energy replenishment data through an optimal algorithm, and determines whether the UAV can complete the unique flight route based on the adjusted estimated energy consumption parameters, and generates a return instruction if it cannot. The mapping module is used to monitor the terrain below the route through video and draw the regional terrain based on the monitoring data; The data transmission module is used to upload the route data and mapping model to the cloud database.

[0013] Preferably, the route acquisition module includes: The input unit is used to input the route start position, route end position and water conservancy surveying and mapping terrain range; The analysis unit is used to analyze the flight routes and mapping objects of the pathways; The generation unit is used to produce complete UAV flight routes.

[0014] Preferably, the surveying and mapping module includes: The camera unit is used to carry out camera monitoring and data storage of the terrain below the route; The rendering unit is used to extract the camera data and render the regional terrain at a reduced scale.

[0015] Preferably, the data transmission module includes: The acquisition unit is used to collect equipment monitoring data and generate data; The pre-processing unit is used to clean, correct and repair the collected data; The transmission unit is used to upload route-related data to the cloud database.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The system of the present invention realizes intelligent planning and management of UAV routes by integrating multiple functional modules. The solar power supply module ensures the continuous operation of the system, the route acquisition and route determination module optimizes the flight route, the surveying and mapping module provides high-precision surveying and mapping data, and the data transmission module ensures the real-time and availability of the data. These modules work together to provide strong technical support for water conservancy surveying and mapping work; The present invention realizes big data planning and management of UAV routes by integrating multiple functions such as solar power supply, route planning, flight control, obstacle avoidance, camera monitoring and terrain mapping. This method not only improves the efficiency and accuracy of surveying and mapping, but also reduces operating costs and environmental impact, providing strong technical support for the development of water conservancy projects. After preprocessing the collected light intensity data and spectral distribution data, they are input into the prediction model to obtain the total photovoltaic power generation in the future scheduled period. Based on the total photovoltaic power generation in the future scheduled period, the maximum range that the UAV can support when executing a single flight route is determined, and the UAV returns when the range is lower than the maximum range to avoid the UAV falling due to lack of power. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the description of the specific implementation methods will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a schematic diagram of the method flow of the present invention; Figure 2 It is a schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Embodiment 1: like Figure 1 As shown, a UAV route big data planning method for water conservancy surveying and mapping includes: S1. Input the starting position of the route, the ending position of the route and the water conservancy surveying and mapping terrain range, and obtain multiple UAV flight routes passing through the water conservancy surveying and mapping area after analysis; S2, determining the estimated consumption distances of multiple UAV flight routes passing through the water conservancy surveying and mapping area and the energy consumption parameters under the estimated consumption distances, and then matching the maximum energy consumption that can be provided by the UAV battery capacity with the energy consumption parameters to give a unique flight route; S3, collecting sunlight intensity data and spectral distribution data in real time to obtain original data; S4, performing preprocessing on the original data to obtain a data set; S5. Input the data set into the prediction model to obtain the total photovoltaic power generation in the future predetermined period; S6. Based on the total amount of photovoltaic power generation in the future scheduled period, the maximum energy consumption is adjusted, and the energy consumption parameters under the only flight route are judged, and an execution strategy is generated based on the judgment result, and uploaded to the drone main control module to generate a corresponding execution command; S7, based on the unique flight route given in step S02, the terrain below the route is monitored and stored in the camera unit during the flight; S8, extracting the camera data by using the drawing unit and drawing the regional terrain at a reduced scale; S9. Upload the route data and surveying model to the cloud database.

[0021] Furthermore, in step S4 of the above embodiment, preprocessing the raw data includes: S41, obtaining a sequence of illumination intensity data and spectral distribution data of raw data; S42, based on the IQR method, detect and mark the abnormal values ​​of the sequence of the light intensity data and the spectral distribution data, the abnormal values ​​are values ​​lower than Q1-1.5*IQR and Q3+1.5*IQR, where: Q1=the 25th percentile value of the sequence of the light intensity data, Q3=the 75th percentile value of the sequence of the spectral distribution data; S43, for the identified abnormal value, select the average value of the first three adjacent data in sequence to replace it, so as to obtain the cleaned light intensity data and the cleaned spectral distribution data; S44, feature extraction of light intensity data after cleaning: S441, calculating the maximum light intensity value in each predetermined time period to obtain a peak value; S442, calculating the average value of multiple light intensities in each predetermined time period; S443, calculating the difference between the maximum value and the minimum value of the light intensity in each predetermined time period to obtain the maximum difference of the light intensity in each predetermined time period; S44. Feature extraction of spectral distribution data after cleaning: S444, selecting a specific wavelength as the light intensity value to obtain the light intensity value at the key wavelength; S445, calculating the integral of the illumination intensity within the entire spectral range to obtain a spectral integral value; S446. Find the wavelength corresponding to the maximum light intensity in the spectral distribution to obtain the peak wavelength value.

[0022] The IQR method effectively identifies and processes outliers in light intensity data and spectral distribution data, improving the accuracy and reliability of the data. This is crucial for subsequent feature extraction and prediction model training, and can significantly reduce the impact of erroneous data on model performance, thereby improving the stability and prediction accuracy of the overall system.

[0023] We further extracted several key features of the light intensity data and spectral distribution data, such as peak value, average value, difference value, light intensity value at key wavelengths, spectral integral value and peak wavelength value, etc. This not only fully reflects the light and spectral characteristics of sunlight, but also provides more abundant and accurate information input for the prediction model, which helps the model to more accurately predict the total amount of photovoltaic power generation in the future.

[0024] Furthermore, the prediction model in step S5 of the above embodiment is generated by defining a prediction model architecture including an input layer, a one-dimensional convolution layer, an activation function layer, a pooling layer, a flattening layer, a fully connected layer and an output layer, wherein: The shape of the input layer matches the dimensions of the spectral data; The one-dimensional convolution layer uses the preset convolution kernel size and number to extract spectral features; The activation function layer uses the ReLU function to increase nonlinearity; The pooling layer uses maximum pooling or average pooling to reduce the feature dimension; The flattening layer flattens the multi-dimensional feature map into a one-dimensional vector; The fully connected layer sets the appropriate number of layers and neurons according to the extracted and cleaned light intensity data features and the extracted and cleaned spectral distribution data features; The output layer is a linear regressor.

[0025] The above utilizes the advantages of convolutional neural networks in feature extraction and pattern recognition, which can efficiently extract useful information from spectral data, and perform nonlinear combination through the fully connected layer, and finally output the predicted value of the total photovoltaic power generation in the future predetermined period.

[0026] Furthermore, in step S5 of the above embodiment, inputting the data set into the prediction model for training includes: S51, inputting the extracted light intensity data features after cleaning and the extracted spectral distribution data features after cleaning into the prediction model, performing forward calculation through the input layer, the one-dimensional convolution layer, the activation function layer, the pooling layer, and the fully connected layer to obtain a predicted output value; S52, calculating the loss value of the predicted output value and the true label based on the mean square error; S53, calculating the gradient of each layer according to the loss value, and transferring the gradient from the output layer to the input layer through the back propagation algorithm; S54, using the optimizer and the calculated gradient to update the weight of the model to obtain a new prediction model, and the predicted output value is used as the true label of the next set of data; wherein the purpose of the weight update is to reduce the loss value; S55. Obtain a forecast data set based on the adjacent differences between the first at least three sets of data as the growth rate, and input it into a new forecast model to obtain the total photovoltaic power generation within a future predetermined period.

[0027] Through iterative training, the model can continuously learn and optimize its internal parameters to more accurately predict the total amount of future photovoltaic power generation. In addition, the process also uses an autoregressive approach to use the predicted output value of the previous round as the true label of the next set of data, further improving the prediction accuracy and stability of the model.

[0028] Furthermore, in step S6 of the above embodiment, the maximum energy consumption is adjusted based on the total amount of photovoltaic power generation in the future predetermined period, and the energy consumption parameter under the only flight route is judged, including: S61, collecting real-time energy consumption of the UAV during operation; S62, collecting the remaining total amount of the current drone battery; S63, based on the obtained total photovoltaic power generation in the future predetermined period and the energy consumption, when the difference is negative, the current remaining amount of the drone battery is used to compensate to zero; S64: When the remaining amount of the current drone battery is equal to the energy consumed by the drone when it has completed the only flight route, the result is determined to be a return flight.

[0029] Based on the predicted total photovoltaic power generation in the future scheduled period, the energy consumption management strategy of the drone is optimized. By real-time monitoring of the drone's energy consumption and the remaining total battery, the system can dynamically adjust the drone's flight plan and energy consumption parameters to ensure that the drone can safely return to the base or recharge while completing the surveying and mapping mission. This not only improves the drone's endurance, but also reduces the risk of mission failure due to insufficient power.

[0030] In summary, after preprocessing the collected light intensity data and spectral distribution data, they are input into the prediction model to obtain the total amount of photovoltaic power generation in the future scheduled period. Based on the total amount of photovoltaic power generation in the future scheduled period, the maximum range that the drone can support when executing a single flight route is determined, and the drone returns home at a range lower than the maximum range (the so-called maximum range refers to the range that the current power is sufficient to support) to avoid the drone crashing due to lack of power.

[0031] Embodiment 2:

[0032] See also Figure 2 As shown, this embodiment is a UAV route big data planning system for water conservancy surveying and mapping, including: A route acquisition module is used to set the starting point and terminal of the target mapping area, generate multiple flight routes within the target mapping area, and provide estimated energy consumption parameters under a unique flight route; Energy storage consumption statistics module, real-time acquisition of actual flight consumption under a unique flight route; Solar power supply module and prediction module, which predicts future photovoltaic energy replenishment data based on the collected light intensity data and spectral distribution data; The data processing module obtains the adjusted estimated energy consumption parameters based on the actual flight consumption and the predicted future photovoltaic energy replenishment data through the optimal algorithm, and determines whether the drone can complete the unique flight route based on the adjusted estimated energy consumption parameters. If not, a return instruction is generated; The mapping module is used to monitor the terrain below the route through video and draw the regional terrain based on the monitoring data; The data transmission module is used to upload the route data and mapping model to the cloud database.

[0033] The above-mentioned route acquisition module includes: The input unit is used to input the route start position, route end position and water conservancy surveying and mapping terrain range; The analysis unit is used to analyze the flight routes and mapping objects of the pathways; The generation unit is used to produce complete UAV flight routes.

[0034] The above-mentioned surveying and mapping module includes: The camera unit is used to carry out camera monitoring and data storage of the terrain below the route; The rendering unit is used to extract the camera data and render the regional terrain at a reduced scale.

[0035] The above-mentioned data transmission module includes: The acquisition unit is used to collect equipment monitoring data and generate data; The pre-processing unit is used to clean, correct and repair the collected data; The transmission unit is used to upload route-related data to the cloud database.

[0036] In this embodiment, the main control module on the UAV serves as the core of the system and is responsible for coordinating and controlling the work of other modules to ensure that the entire route planning and water conservancy surveying and mapping process can proceed smoothly and orderly. The solar power supply module uses solar energy for energy conversion to provide the required power for the UAV system. The use of this green energy helps to extend the flight time of the UAV and reduce dependence on traditional energy. The route acquisition module calculates and generates multiple possible flight routes in the target area based on the input route start and end positions and surveying area information, which provides basic data for subsequent route optimization. The route determination module selects the optimal route from multiple possible flight routes through comparative analysis, such as considering factors such as flight time, energy consumption, and safety, which helps the UAV complete the surveying and mapping task in a more efficient and safer way.

[0037] The flight module controls the flight attitude and trajectory of the drone through the cooperation of propellers and servos, which is the key to the drone's ability to fly according to the predetermined route and achieve mapping tasks. During the flight, the obstacle avoidance module senses and identifies abnormal obstacles on the route in real time, such as trees and buildings, and takes corresponding avoidance measures, which helps to ensure the flight safety of the drone. The mapping module uses camera equipment to monitor the terrain below the route and draws the regional terrain based on the collected data. This is one of the core functions of drones in water conservancy mapping. The data transmission module uploads key information such as route data and mapping models collected by the drone to the cloud database for storage and management, which helps to achieve real-time data sharing and remote access, and facilitates subsequent analysis and processing. By integrating multiple functional modules, this system realizes the intelligent planning and management of UAV routes. The solar power supply module ensures the continuous operation of the system, the route acquisition and route determination module optimizes the flight route, the surveying and mapping module provides high-precision surveying and mapping data, and the data transmission module ensures the real-time and availability of the data. These modules work together to provide strong technical support for water conservancy surveying and mapping work.

[0038] During the flight, the obstacle avoidance module can sense and identify abnormal obstacles on the route, such as trees and buildings, in real time, and take corresponding avoidance measures, such as adjusting the flight altitude, direction or speed, to ensure the safety of the drone.

[0039] During the flight, the camera unit will monitor the terrain below the route and store data, which will be used for subsequent terrain mapping and surveying and mapping analysis.

[0040] The drawing unit is used to extract the camera data and draw the regional terrain at a reduced scale. The drawing result will present key information such as the topographic features and water flow conditions of the terrain.

[0041] The route data and surveying and mapping models are uploaded to the cloud database. Users can access the cloud database through mobile terminals, computers and other devices to view and manage the surveying and mapping data. At the same time, they can also use data analysis tools to conduct in-depth mining and analysis of the surveying and mapping data to provide a scientific basis for the planning, design and management of water conservancy projects.

[0042] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0043] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A UAV route big data planning method for water conservancy surveying and mapping, characterized in that: The following steps are involved: S1. Input the starting position of the route, the ending position of the route and the water conservancy surveying and mapping terrain range, and obtain multiple UAV flight routes passing through the water conservancy surveying and mapping area after analysis; S2, determining the estimated consumption distances of multiple UAV flight routes passing through the water conservancy surveying and mapping area and the energy consumption parameters under the estimated consumption distances, and then matching the maximum energy consumption that can be provided by the UAV battery capacity with the energy consumption parameters to give a unique flight route; S3, collecting sunlight intensity data and spectral distribution data in real time to obtain original data; S4, performing preprocessing on the original data to obtain a data set; S5. Input the data set into the prediction model to obtain the total photovoltaic power generation in the future predetermined period; S6, adjusting the maximum energy consumption based on the total amount of photovoltaic power generation in the future predetermined period, and judging the energy consumption parameters under the only flight route, generating an execution strategy based on the judgment result, and uploading it to the drone main control module to generate a corresponding execution command; S7, based on the unique flight route given in step S02, the terrain below the route is monitored and stored in the camera unit during the flight; S8, extracting the camera data by using the drawing unit and drawing the regional terrain at a reduced scale; S9. Upload the route data and surveying model to the cloud database.

2. The method for planning UAV route big data for water conservancy surveying and mapping according to claim 1 is characterized by: The preprocessing of the raw data in step S4 includes: S41, obtaining a sequence of light intensity data and spectral distribution data of the original data; S42, based on the IQR method, respectively detect and mark the abnormal values ​​of the sequence of the light intensity data and the spectral distribution data, wherein the abnormal values ​​are values ​​lower than Q1-1.5*IQR and Q3+1.5*IQR, wherein: Q1=the 25th percentile value of the sequence of the light intensity data, Q3=the 75th percentile value of the sequence of the spectral distribution data; S43, for the identified abnormal value, select the average value of the first three adjacent data in sequence to replace it, so as to obtain the cleaned light intensity data and the cleaned spectral distribution data.

3. The method for planning UAV route big data for water conservancy surveying and mapping according to claim 2 is characterized by: The step S4 of preprocessing the obtained cleaned light intensity data and cleaned spectral distribution data further includes: S44, feature extraction of light intensity data after cleaning: S441, calculating the maximum light intensity value in each predetermined time period to obtain a peak value; S442, calculating the average value of multiple light intensities in each predetermined time period; S443, calculating the difference between the maximum value and the minimum value of the light intensity in each predetermined time period to obtain the maximum difference of the light intensity in each predetermined time period; S44. Feature extraction of spectral distribution data after cleaning: S444, selecting a specific wavelength as the light intensity value to obtain the light intensity value at the key wavelength; S445, calculating the integral of the illumination intensity within the entire spectral range to obtain a spectral integral value; S446. Find the wavelength corresponding to the maximum light intensity in the spectral distribution to obtain the peak wavelength value.

4. The method for planning UAV route big data for water conservancy surveying and mapping according to claim 1 is characterized by: The prediction model in step S5 is generated by defining a prediction model architecture including an input layer, a one-dimensional convolution layer, an activation function layer, a pooling layer, a flattening layer, a fully connected layer and an output layer, wherein: The shape of the input layer matches the dimensions of the spectral data; The one-dimensional convolution layer uses the preset convolution kernel size and number to extract spectral features; The activation function layer uses the ReLU function to increase nonlinearity; The pooling layer uses maximum pooling or average pooling to reduce the feature dimension; The flattening layer flattens the multi-dimensional feature map into a one-dimensional vector; The fully connected layer sets the appropriate number of layers and neurons according to the extracted and cleaned light intensity data features and the extracted and cleaned spectral distribution data features; The output layer is a linear regressor.

5. The method for planning UAV route big data for water conservancy surveying and mapping according to claim 4 is characterized by: In step S5, inputting the data set into the prediction model for training includes: S51, inputting the extracted light intensity data features after cleaning and the extracted spectral distribution data features after cleaning into the prediction model, performing forward calculation through the input layer, the one-dimensional convolution layer, the activation function layer, the pooling layer, and the fully connected layer to obtain a predicted output value; S52, calculating the loss value of the predicted output value and the true label based on the mean square error; S53, calculating the gradient of each layer according to the loss value, and transferring the gradient from the output layer to the input layer through the back propagation algorithm; S54, using the optimizer and the calculated gradient to update the weight of the model to obtain a new prediction model, and the predicted output value is used as the true label of the next set of data; wherein the purpose of the weight update is to reduce the loss value; S55. Obtain a forecast data set based on adjacent differences between at least three sets of data as the growth rate, and input it into the new forecast model to obtain the total photovoltaic power generation within a future predetermined period.

6. The method for planning UAV route big data for water conservancy surveying and mapping according to claim 1 is characterized by: In step S6, adjusting the maximum energy consumption based on the total amount of photovoltaic power generation in the future predetermined period, and determining the energy consumption parameter under the only flight route include: S61, collecting real-time energy consumption of the UAV during operation; S62, collecting the remaining total amount of the current drone battery; S63, based on the obtained total photovoltaic power generation in the future predetermined period and the energy consumption, when the difference is a negative value, the remaining total amount of the current drone battery is used to compensate to zero; S64: When the remaining amount of the current drone battery is equal to the energy consumed by the drone after completing the unique flight route, the result of the determination is return flight.

7. A UAV route big data planning system for water conservancy surveying and mapping, which is used to implement the UAV route big data planning method for water conservancy surveying and mapping as described in any one of claims 1 to 6, characterized in that: include: A route acquisition module, which is used to set the starting point and the terminal of the target mapping area, generate multiple flight routes within the target mapping area, and provide estimated energy consumption parameters under a unique flight route; Energy storage consumption statistics module, real-time acquisition of actual flight consumption under a unique flight route; Solar power supply module and prediction module, which predicts future photovoltaic energy replenishment data based on the collected light intensity data and spectral distribution data; A data processing module, which obtains the adjusted estimated energy consumption parameters based on the actual flight consumption and the predicted future photovoltaic energy replenishment data through an optimal algorithm, and determines whether the UAV can complete the unique flight route based on the adjusted estimated energy consumption parameters, and generates a return instruction if it cannot. The mapping module is used to monitor the terrain below the route through video and draw the regional terrain based on the monitoring data; The data transmission module is used to upload the route data and mapping model to the cloud database.

8. The UAV route big data planning system for water conservancy surveying and mapping according to claim 7 is characterized by: The route acquisition module includes: The input unit is used to input the route start position, route end position and water conservancy surveying and mapping terrain range; The analysis unit is used to analyze the flight routes and mapping objects of the pathways; The generation unit is used to produce complete UAV flight routes.

9. The UAV route big data planning system for water conservancy surveying and mapping according to claim 7 is characterized by: The surveying and mapping module comprises: The camera unit is used to carry out camera monitoring and data storage of the terrain below the route; The rendering unit is used to extract the camera data and render the regional terrain at a reduced scale.

10. The UAV route big data planning system for water conservancy surveying and mapping according to claim 7 is characterized by: The data transmission module comprises: The acquisition unit is used to collect equipment monitoring data and generate data; The pre-processing unit is used to clean, correct and repair the collected data; The transmission unit is used to upload route-related data to the cloud database.

Images