Intelligent three-dimensional management and control platform based on low-altitude miniature airport weather and airspace guarantee

Abstract

The invention discloses an intelligent three-dimensional management and control platform based on low-altitude miniature airport meteorology and airspace guarantee, and belongs to the technical field of intelligent traffic. An intelligent three-dimensional management and control platform based on low-altitude miniature airport meteorology and airspace guarantee comprises a meteorological data processing module used for obtaining various meteorological data, carrying out data fusion on the various meteorological data, and carrying out meteorological prediction of an unmanned aerial vehicle airport by using the fused meteorological data; and the airspace management module is used for determining the flight path of the unmanned aerial vehicle based on the meteorological prediction result and the airspace limitation. According to the invention, the accuracy of airport weather prediction and the reliability of flight path planning are improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation technology, and in particular relates to an intelligent three-dimensional management and control platform based on low-altitude micro-airport meteorological and airspace security. Background Art

[0002] With the rapid development of the low-altitude economy, drones are being used more and more widely in various fields. Explore the drone control system, build a new, multi-dimensional, high-tech three-dimensional flight transportation network and supervision and control platform that integrates high and low, covers lines and surfaces, as an effective supplement to the low-altitude air traffic control service guarantee system and airport network. The existing drone control platform has problems such as insufficient information interaction, insufficient data processing capabilities, and imperfect safety protection measures, which makes it difficult to meet the needs of drone operations in complex environments. In order to transform the negative interference factors of drones into safe tools and efficient elements, an intelligent and three-dimensional control algorithm is urgently needed to improve operational efficiency and safety. The realization of connecting peer networks into a surface and building a full-coverage and dead-end-free delivery and transportation network can greatly change traditional production, sales, consumption and life patterns, and greatly promote economic development. Summary of the invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide an intelligent three-dimensional management and control platform based on low-altitude micro-airport meteorological and airspace security.

[0004] The purpose of the present invention is achieved through the following technical solutions: an intelligent three-dimensional control platform based on low-altitude micro-airport meteorology and airspace security, including:

[0005] The meteorological data processing module is used to obtain various meteorological data, fuse the various meteorological data, and use the fused meteorological data to make weather forecasts for drone airports;

[0006] The airspace management module is used to determine the flight path of the drone based on the results of weather forecasts and airspace restrictions.

[0007] Furthermore, the meteorological data processing module includes:

[0008] Meteorological data acquisition module, used to obtain various meteorological data;

[0009] Meteorological data fusion module, used to identify and correct abnormal data in meteorological data, fill in missing data in meteorological data, and convert multiple meteorological data into a unified format so that multiple meteorological data are consistent in timestamps;

[0010] The meteorological forecast module is used to extract the spatial feature vectors of meteorological data using convolutional neural networks; based on the spatial feature vectors of meteorological data, multiple decision trees in the random forest model are used to independently perform meteorological forecasts; the Stacking strategy module is used to stack multiple prediction results of the random forest model into a fused feature vector, and the final meteorological forecast result is generated based on the fused feature vector.

[0011] Furthermore, the meteorological data fusion module is specifically used for:

[0012] Statistical discrimination method is used to identify abnormal data in meteorological data, and linear regression model is used to correct abnormal data in meteorological data;

[0013] The missing data in the meteorological data are filled in, and the calculation formula for the filled data is:

[0014]

[0015] In the formula, Indicates filling data; Represents the data value of the kth neighboring point; Represents the weight value of the kth neighboring point. The closer to the point where data is missing, the larger the weight value.

[0016] Convert various meteorological data into a unified format to make the various meteorological data consistent in time stamps.

[0017] Furthermore, the airspace management module includes:

[0018] A graph model building module, used to build a graph model according to the target airspace;

[0019] The path generation module is used to determine the flight path of the drone based on the results of weather forecasts, airspace restrictions and graph models.

[0020] Furthermore, the graph model construction module is specifically used for:

[0021] Divide the target airspace into multiple grids, each grid as a node, and each node contains location information and meteorological information;

[0022] Connect adjacent nodes to form edges, forming a graph model;

[0023] Determine the fixed weight of each edge in the graph model. The calculation formula of the fixed weight is:

[0024]

[0025] In the formula, represents the fixed weight of edge Q between node a and node b, represents the distance between node a and node b, Represented as an integer to avoid division by 0. represents the frequency of the paths from nodes a and b being selected in the historical flight paths, represents a reference frequency, Indicates the average delay time from node a and node b in the historical delay records. , and They represent the weight indexes of distance, historical flight path, and historical delay record respectively.

[0026] Further, the path generation module is specifically used to execute step S100 to step S900;

[0027] Step S100. Determine the starting node and the target node;

[0028] Step S200. Initialize the distance array, the predecessor node array and the access mark array; the distance array is used to store the shortest distance estimate from the starting node to each node in the graph model, the distance of the starting node is 0 initially, and the distances of the remaining nodes are infinite; the predecessor node array is used to record the predecessor node of each node on the shortest path, and the predecessor node array is empty initially; the access mark array is used to mark whether the nodes in the graph model have been visited, and all nodes are marked as unvisited initially;

[0029] Step S300. Determine the dynamic weight of each edge in the graph model according to the dynamic weight function, where the dynamic weight function is:

[0030]

[0031] In the formula, represents the dynamic weight of edge Q at time t, represents the dynamic weighted impact of the ith meteorological condition on edge Q at time t, represents the weight coefficient of the i-th meteorological condition, represents the severity of the i-th meteorological condition of edge Q at time t (such as wind speed, precipitation intensity, etc.), represents the impact of spatial constraints on the dynamic weight of edge Q at time t, , and They represent the weight coefficients of airspace restrictions when edge P is located in the no-fly zone, restricted-fly zone, and ordinary airspace, respectively. represents the route congestion degree of edge Q at time t;

[0032] Step S400. Determine the overall weight of each edge according to the fixed weight and dynamic weight of each edge;

[0033] Step S500. Find the unvisited node with the smallest distance in the distance array, and mark the node as the current node;

[0034] Step S600. Mark the current node as visited in the visit mark array;

[0035] Step S700. Calculate the distance from the current node to each adjacent node respectively. If the distance is less than the distance of the adjacent node in the distance array, update the distance of the adjacent node in the distance array to the new distance, and record the predecessor node of the adjacent node as the current node in the predecessor node array;

[0036] Step S800. Determine whether the target node is found. If the target node is found, execute step S900, otherwise execute step S500;

[0037] Step S900: Based on the predecessor node array, trace back from the target node to the start node to construct the shortest path.

[0038] Furthermore, the airspace management module also includes a ground path planning module, which is used to plan the shortest path for ground transportation based on the flight path of the drone using an A* search algorithm.

[0039] Furthermore, the intelligent three-dimensional integrated management and control platform also includes a UAV charging and discharging management module, which is used to determine the charging strategy of the UAV's battery and monitor the UAV's battery.

[0040] Furthermore, the UAV charging and discharging management module includes:

[0041] The charging and discharging scheduling module is used to predict the charging and discharging trend of the battery based on the flight plan, battery capacity and current charging status of the UAV using an integrated exponential smoothing prediction algorithm, and to determine the charging priority of the UAV using a charging scheduling algorithm based on a priority queue;

[0042] The battery detection module is used to obtain the performance data of the drone and monitor the remaining power and health status of the battery based on the state estimation algorithm of the Kalman filter;

[0043] The battery test module is used to regularly perform open circuit voltage and capacity tests on the drone's batteries, and identify battery failure modes and maintenance requirements through statistical FMFA analysis methods.

[0044] The beneficial effects of the present invention are:

[0045] (1) The present invention integrates multiple meteorological data to perform weather forecasts for drone airports, thereby improving the accuracy of airport weather forecasts and making it more suitable for weather forecasts at low-altitude micro-airports;

[0046] (2) The present invention plans the flight path of the UAV based on the airport's weather forecast results and airspace restrictions, thereby improving the reliability of flight path planning;

[0047] (3) The present invention adopts a scheduling strategy based on a priority queue, dynamically adjusts the charging and discharging sequence in combination with the UAV's flight plan, battery capacity, and current charging status, thereby achieving efficient management and optimized allocation of battery energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A block diagram of the intelligent three-dimensional integrated management and control platform in the present invention;

[0049] Figure 2 A flow chart for generating the flight path of the UAV in the present invention. DETAILED DESCRIPTION

[0050] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0051] See also Figure 1 to Figure 2 The present invention provides an intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace security:

[0052] like Figure 1 As shown, the intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace support includes a meteorological data processing module and an airspace management module. The meteorological data processing module is used to obtain a variety of meteorological data, fuse the multiple meteorological data, and use the fused meteorological data to perform meteorological forecasts for drone airports; the airspace management module is used to determine the flight path of drones based on the results of meteorological forecasts and airspace restrictions.

[0053] In this embodiment, a variety of meteorological data are integrated and then a weather forecast is performed, which improves the accuracy of the weather forecast for the drone airport; the flight path of the drone is planned based on the weather forecast results and airspace restrictions, which reduces the flight time and other flight costs of the drone and avoids air traffic congestion.

[0054] In some implementations of this embodiment, the meteorological data processing module includes a meteorological data acquisition module, a meteorological data fusion module and a meteorological prediction module.

[0055] The meteorological data acquisition module is used to acquire various meteorological data. For example, the meteorological data acquired by the meteorological data acquisition module includes ground meteorological station data, multi-band meteorological radar data, satellite remote sensing data, and output data of numerical weather forecast models.

[0056] The meteorological data fusion module is used to fuse multiple meteorological data.

[0057] Specifically, multiple meteorological data are integrated, including: identifying and correcting abnormal data in meteorological data; filling in missing data in meteorological data; converting multiple meteorological data into a unified format to make the multiple meteorological data consistent in time stamps.

[0058] In these implementations, by unifying the formats of various meteorological data, it is ensured that all datasets of meteorological data are consistent in timestamps.

[0059] In some implementations of this embodiment, the identification and correction of abnormal data in meteorological data include: using a statistical discrimination method to identify abnormal data in meteorological data; and using a linear regression model to correct the abnormal data in meteorological data.

[0060] In these implementations, abnormal data is identified and corrected, thereby improving the quality and reliability of meteorological data and the reliability of meteorological forecast results.

[0061] In some implementations of this embodiment, when filling in missing data in meteorological data, the calculation formula for filling in the data is:

[0062]

[0063] In the formula, Indicates filling data; Represents the data value of the kth neighboring point; Represents the weight value of the kth neighboring point. The closer to the point where data is missing, the larger the weight value.

[0064] In these implementations, missing data are filled in, so as to maintain the temporal integrity of meteorological data and facilitate subsequent meteorological forecasting.

[0065] The weather forecast module is used to use the fused weather data to perform weather forecasts for drone airports.

[0066] In some implementations of this embodiment, the weather forecast module includes a convolutional neural network, a random forest model and a Stacking strategy module.

[0067] The convolutional neural network is used to extract the spatial feature vector of meteorological data. The input of the convolutional neural network is a numerical matrix of meteorological data, in which each row represents a meteorological observation value at a time point, and each column represents a different meteorological variable, such as temperature, humidity, wind speed, etc. For spatial information such as radar images, the radar image data is directly used as the input of the convolutional neural network.

[0068] Specifically, the convolutional neural network includes a convolutional layer, an activation layer, a pooling layer, and a fully connected layer. The convolutional layer is used to extract the spatial features of meteorological data; the activation layer is used to increase nonlinearity and improve the expression ability of the model. In this embodiment, the ReLU activation function is used; the pooling layer is used to reduce the dimension of the feature map, reduce the amount of calculation, and improve the robustness of the feature; the fully connected layer is used to integrate the features extracted by the convolutional layer into a fixed-length spatial feature vector for subsequent prediction.

[0069] The random forest model is used to independently perform weather forecasting using multiple decision trees based on the spatial feature vectors of weather data.

[0070] The construction process of the random forest model is as follows: construct multiple decision trees, and train each decision tree based on the input feature vector, that is, the training of each decision tree is independent; in the training process of the random forest model, random sampling and feature selection methods are used to increase the diversity and robustness of the random forest model. When performing weather forecasts, for each input spatial feature vector, the random forest model uses multiple decision trees to give corresponding prediction results, where the predicted values ​​of meteorological variables (such as temperature, humidity, etc.) are output for regression tasks, and classification results (such as weather type, rainfall probability, etc.) are output for classification tasks.

[0071] In some implementations of this embodiment, the parameters of the random forest model are dynamically adjusted based on new observation data and feedback on prediction accuracy to adapt to rapidly changing meteorological conditions.

[0072] The Stacking strategy module is used to integrate multiple prediction results of the random forest model to obtain the final meteorological forecast result.

[0073] Specifically, the Stacking strategy module stacks multiple prediction results of the random forest model into a fused feature vector, and generates a final meteorological forecast result based on the fused feature vector.

[0074] In these implementations, a convolutional neural network is used to extract spatial feature vectors of meteorological data, and then a random forest model is used to make preliminary predictions on these spatial feature vectors. Finally, multiple prediction results of the random forest model are integrated through a stacking strategy module, thereby improving the accuracy and robustness of meteorological prediction results for drone airports.

[0075] In some implementations of this embodiment, the weather forecast results include short-term weather forecasts, which include weather forecasts within 0-2 hours (visibility, cloud height, wind speed and direction, and precipitation conditions), and disastrous weather forecasts (wind shear, hail, downbursts, turbulence, and sandstorms), thereby ensuring the safe operation of drones.

[0076] In some implementations of this embodiment, the airspace management module includes a graph model construction module and a path generation module.

[0077] The graph model construction module is used to construct a graph model according to the target airspace.

[0078] Specifically, the construction process of the graph model is as follows: divide the target airspace into multiple grids, each grid as a node, and each node contains location information (such as latitude, longitude and altitude) and meteorological information (such as wind speed, wind direction, etc.). Connect adjacent nodes to form edges to form a graph model. Determine the fixed weight of each edge in the model graph. The calculation formula for the fixed weight is:

[0079]

[0080] In the formula, represents the fixed weight of edge Q between node a and node b, represents the distance between node a and node b, Represented as an integer to avoid division by 0. represents the frequency of the paths from nodes a and b being selected in the historical flight paths, represents a reference frequency, Indicates the average delay time from node a and node b in the historical delay records. , and They represent the weight indexes of distance, historical flight path, and historical delay record respectively.

[0081] In these implementations, the fixed weight of each edge is calculated based on the distance, historical flight path, and historical delay records, which can more accurately reflect the quality of the actual flight path.

[0082] In some implementations of this embodiment, the fixed weight of each edge in the graph model is updated periodically.

[0083] The path generation module is used to determine the flight path of the UAV based on the results of weather forecasts, airspace restrictions and graph models.

[0084] Specifically, Figure 2As shown, the method for determining the flight path of the UAV includes steps S100 to S900.

[0085] Step S100: Determine the starting node and the target node.

[0086] Step S200. Initialize the distance array, the predecessor node array, and the access mark array. The distance array is used to store the shortest distance estimate from the starting node to each node in the graph model. Initially, the distance of the starting node is 0, and the distances of the remaining nodes are infinite; the predecessor node array is used to record the predecessor node of each node on the shortest path. Initially, the predecessor node array is empty; the access mark array is used to mark whether the nodes in the graph model have been visited. Initially, all nodes are marked as unvisited.

[0087] Step S300. Determine the dynamic weight of each edge in the graph model according to the dynamic weight function, where the dynamic weight function is:

[0088]

[0089] In the formula, represents the dynamic weight of edge Q at time t, represents the dynamic weighted impact of the ith meteorological condition on edge Q at time t, represents the weight coefficient of the i-th meteorological condition, represents the severity of the i-th meteorological condition of edge Q at time t (such as wind speed, precipitation intensity, etc.), represents the impact of spatial constraints on the dynamic weight of edge Q at time t, , and They represent the weight coefficients of airspace restrictions when edge P is located in the no-fly zone, restricted-fly zone, and ordinary airspace, respectively. It represents the route congestion degree of edge Q at time t.

[0090] Step S400: Determine the overall weight of each edge according to the fixed weight and dynamic weight of each edge.

[0091] For example, the overall weight of each edge is the sum of its corresponding fixed weight and dynamic weight.

[0092] Step S500: Find the unvisited node with the shortest distance in the distance array, and mark the node as the current node.

[0093] Step S600: Mark the current node as visited in the visit mark array.

[0094] Step S700. Calculate the distance from the current node to each adjacent node respectively. If the distance is less than the distance of the adjacent node in the distance array, update the distance of the adjacent node in the distance array to the new distance, and record the predecessor node of the adjacent node in the predecessor node array as the current node.

[0095] Step S800: Determine whether the target node is found. If the target node is found, execute step S900; otherwise, execute step S500.

[0096] Step S900: Based on the predecessor node array, trace back from the target node to the start node to construct the shortest path.

[0097] In these implementations, the weights of each edge in the graph model are dynamically adjusted based on meteorological information and airspace restrictions, which can generate a better flight path for the drone and improve the efficiency and safety of air traffic.

[0098] In some implementations of this embodiment, the flight area is determined based on the determined flight path of the UAV, and the flight plan (flight path, flight altitude, pitch angle, etc.) of the UAV is determined based on the flight environment (altitude, topography, wind direction and speed) and weather condition assessment (cloud thickness, light intensity, air visibility) of the flight area, thereby promoting the rational use of airspace resource management and ensuring flight safety.

[0099] In some implementations of this embodiment, the airspace management module also includes a ground path planning module, which is used to plan the shortest path for ground transportation based on the flight path of the drone using an A* search algorithm.

[0100] In these embodiments, the shortest path is planned for ground vehicles based on the flight path of the drone to address potential ground-air coordination issues.

[0101] In some implementations of this embodiment, the intelligent three-dimensional integrated management and control platform also includes a drone charging and discharging management module, which is used to determine the charging strategy of the drone's battery and monitor the drone's battery.

[0102] In some implementations of this embodiment, the drone charge and discharge management module includes a charge and discharge scheduling module, a battery detection module and a battery testing module.

[0103] The charging and discharging scheduling module is used to optimize the charging and discharging decision by using an integrated exponential smoothing state prediction algorithm and heuristic rules through a scheduling strategy based on a priority queue, combined with the aircraft flight plan, battery capacity and current charging state, so as to achieve efficient management and optimized allocation of battery energy.

[0104] The battery detection module is used to obtain the performance data of the drone (such as voltage, current, charge and discharge cycle and environmental conditions, etc.), and monitor the remaining power and health status of the battery based on the state estimation algorithm of Kalman filtering.

[0105] The battery testing module is used to regularly perform open circuit voltage test and capacity test on the battery of the drone, and identify the failure mode and maintenance requirements of the battery through the statistical FMFA analysis method.

[0106] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. An intelligent three-dimensional control platform based on low-altitude micro-airport meteorology and airspace security, characterized by: include: The meteorological data processing module is used to obtain various meteorological data, fuse the various meteorological data, and use the fused meteorological data to make weather forecasts for drone airports; An airspace management module, used to determine the flight path of the drone based on the results of weather forecasts and airspace restrictions; The meteorological data processing module comprises: Meteorological data acquisition module, used to obtain various meteorological data; Meteorological data fusion module, used to identify and correct abnormal data in meteorological data, fill in missing data in meteorological data, and convert multiple meteorological data into a unified format so that multiple meteorological data are consistent in timestamps; The meteorological forecast module is used to extract the spatial feature vectors of meteorological data using convolutional neural networks; based on the spatial feature vectors of meteorological data, multiple decision trees in the random forest model are used to independently perform meteorological forecasts; the Stacking strategy module is used to stack multiple prediction results of the random forest model into a fused feature vector, and the final meteorological forecast result is generated based on the fused feature vector.

2. The intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace security according to claim 1 is characterized in that: The meteorological data fusion module is specifically used for: Statistical discrimination method is used to identify abnormal data in meteorological data, and linear regression model is used to correct abnormal data in meteorological data; The missing data in the meteorological data are filled in, and the calculation formula for the filled data is: ； In the formula, Indicates filling data; Represents the data value of the kth neighboring point; Represents the weight value of the kth neighboring point. The closer to the point where data is missing, the larger the weight value. Convert various meteorological data into a unified format to make the various meteorological data consistent in time stamps.

3. The intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace security according to claim 1 is characterized in that: The airspace management module includes: A graph model building module, used to build a graph model according to the target airspace; The path generation module is used to determine the flight path of the drone based on the results of weather forecasts, airspace restrictions and graph models.

4. The intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace security according to claim 3 is characterized in that: The graph model building module is specifically used for: Divide the target airspace into multiple grids, each grid as a node, and each node contains location information and meteorological information; Connect adjacent nodes to form edges, forming a graph model; Determine the fixed weight of each edge in the graph model. The calculation formula of the fixed weight is: ； In the formula, represents the fixed weight of edge Q between node a and node b, represents the distance between node a and node b, Represented as an integer to avoid division by 0. represents the frequency of the paths from nodes a and b being selected in the historical flight paths, represents a reference frequency, Indicates the average delay time from node a and node b in the historical delay records. , and They represent the weight indexes of distance, historical flight path, and historical delay record respectively.

5. The intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace security according to claim 3 is characterized in that: The path generation module is specifically used to execute steps S100 to S900; Step S100. Determine the starting node and the target node; Step S200. Initialize the distance array, the predecessor node array and the access mark array; the distance array is used to store the shortest distance estimate from the starting node to each node in the graph model, the distance of the starting node is 0 initially, and the distances of the remaining nodes are infinite; the predecessor node array is used to record the predecessor node of each node on the shortest path, and the predecessor node array is empty initially; the access mark array is used to mark whether the nodes in the graph model have been visited, and all nodes are marked as unvisited initially; Step S300. Determine the dynamic weight of each edge in the graph model according to the dynamic weight function, where the dynamic weight function is: ； In the formula, represents the dynamic weight of edge Q at time t, represents the dynamic weighted impact of the ith meteorological condition on edge Q at time t, represents the weight coefficient of the i-th meteorological condition, represents the severity of the i-th meteorological condition of edge Q at time t, represents the impact of spatial constraints on the dynamic weight of edge Q at time t, , and They represent the weight coefficients of airspace restrictions when edge P is located in the no-fly zone, restricted-fly zone, and ordinary airspace, respectively. represents the route congestion degree of edge Q at time t; Step S400. Determine the overall weight of each edge according to the fixed weight and dynamic weight of each edge; Step S500. Find the unvisited node with the smallest distance in the distance array, and mark the node as the current node; Step S600. Mark the current node as visited in the visit mark array; Step S700. Calculate the distance from the current node to each adjacent node respectively. If the distance is less than the distance of the adjacent node in the distance array, update the distance of the adjacent node in the distance array to the new distance, and record the predecessor node of the adjacent node as the current node in the predecessor node array; Step S800. Determine whether the target node is found. If the target node is found, execute step S900, otherwise execute step S500; Step S900: Based on the predecessor node array, trace back from the target node to the start node to construct the shortest path.

6. The intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace security according to claim 3 is characterized in that: The airspace management module also includes a ground path planning module, which is used to plan the shortest path for ground transportation vehicles based on the flight path of the drone using an A* search algorithm.

7. The intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace security according to claim 1 is characterized in that: The intelligent three-dimensional integrated management and control platform also includes a UAV charging and discharging management module, which is used to determine the charging strategy of the UAV's battery and monitor the UAV's battery.

8. The intelligent three-dimensional control platform based on low-altitude micro-airport meteorological and airspace security according to claim 7 is characterized in that: The UAV charging and discharging management module includes: The charging and discharging scheduling module is used to predict the charging and discharging trend of the battery based on the flight plan, battery capacity and current charging status of the UAV using an integrated exponential smoothing prediction algorithm, and to determine the charging priority of the UAV using a charging scheduling algorithm based on a priority queue; The battery detection module is used to obtain the performance data of the drone and monitor the remaining power and health status of the battery based on the state estimation algorithm of the Kalman filter; The battery test module is used to regularly perform open circuit voltage and capacity tests on the drone's batteries, and identify battery failure modes and maintenance requirements through statistical FMFA analysis methods.

Images