Unmanned aerial vehicle management method and system based on big data and program product
The drone management method, which combines big data matching and multi-indicator feature fusion, addresses the shortcomings in trajectory planning, anomaly detection, and control in drone management, and achieves an efficient and reliable closed-loop drone management system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-03
- Publication Date
- 2026-03-10
AI Technical Summary
Existing drone management methods are inadequate in terms of trajectory planning adaptability, anomaly detection accuracy, and differentiated anomaly handling, making it difficult to adapt to the needs of complex flight missions.
By acquiring flight mission data from drones and performing big data matching, flight trajectories are planned, and flight status and position parameters are collected in real time. Multi-indicator feature fusion and anomaly detection are performed to output early warning information and control commands to achieve differentiated management and control.
It has achieved efficient drone trajectory planning, improved the timeliness and accuracy of flight anomaly identification, realized reliable flight status perception and differentiated control, and formed a complete drone management closed loop.
Smart Images
Figure CN121635007A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of drone management technology, specifically relating to drone management methods, systems, and program products based on big data. Background Technology
[0002] Unmanned aerial vehicles (UAVs) are unmanned aerial vehicles controlled by radio remote control equipment and their own program control system. They have advantages such as small size, low cost, and ease of use. With the rapid development of UAV technology, UAVs have been widely used in many fields such as logistics delivery, agricultural plant protection, power line inspection, and aerial photography and mapping.
[0003] However, the surge in the number of drones and the increasing complexity of their application scenarios have brought numerous challenges to drone management, exposing shortcomings in existing drone control methods. These shortcomings include: 1. Insufficient adaptability of trajectory planning, making it difficult to adapt to different flight mission requirements. 2. Anomaly detection during flight is mostly based on a single data source or simple threshold judgment, which is insufficient to cope with complex flight scenarios and lacks accuracy in anomaly detection. 3. Anomaly handling often adopts a one-size-fits-all strategy, lacking differentiated control measures for different anomaly determinations. Summary of the Invention
[0004] The purpose of this invention is to provide a method, system, and program product for managing drones based on big data, in order to solve the above-mentioned problems existing in the prior art.
[0005] To achieve the above objectives, the present invention adopts the following technical solution: Firstly, it provides a big data-based approach to drone management, including: Acquire the flight mission data of the target drone, and perform big data matching of drone flight missions in the target airspace based on the flight mission data of the target drone to determine the planned flight trajectory of the target drone in the target airspace; The flight control data of the target UAV is generated based on the planned flight trajectory of the target UAV in the target airspace, and the flight control data is sent to the target UAV so that the target UAV can perform flight missions in the target airspace according to the flight control data; During the flight mission of the target UAV within the target airspace, the flight status parameters and flight position parameters of the target UAV are collected in real time. Flight status is assessed based on flight status parameters, and status index features are extracted. Flight trajectory is assessed based on flight position parameters and planned flight trajectory, and trajectory index features are extracted. Feature fusion is performed on state indicator features and trajectory indicator features to obtain multi-indicator joint features; The joint features of multiple indicators are input into a pre-set joint anomaly detection model to perform anomaly detection and obtain joint anomaly detection results. When the target UAV is determined to have a joint anomaly based on the joint anomaly detection results, the first anomaly warning information of the target UAV is output, and an emergency landing control command is sent to the target UAV.
[0006] In one possible design, the step of performing big data matching of UAV flight missions in the target airspace based on the flight mission data of the target UAV to determine the planned flight trajectory of the target UAV within the target airspace includes: Determine the starting and ending waypoints of the target UAV within the target airspace based on the flight mission data of the target UAV. Search and match all historical flight mission trajectories from the starting waypoint to the ending waypoint in the big data of UAV flight missions in the target airspace, and determine the currently unoccupied historical flight mission trajectories based on the big data of UAV flight missions in the target airspace, and use the currently unoccupied historical flight mission trajectories as pre-selected flight trajectories; The shortest flight path among all pre-selected flight paths is chosen as the planned flight path for the target UAV within the target airspace.
[0007] In one possible design, the flight state parameters include flight speed parameters, flight attitude parameters, airframe vibration parameters, battery health parameters, and motor load parameters; the flight position parameters include flight longitude parameters, flight latitude parameters, and flight altitude parameters; and the planned flight trajectory is a four-dimensional flight trajectory that includes planned longitude data, planned latitude data, planned altitude data, and planned time data.
[0008] In one possible design, the flight state assessment based on flight state parameters and the extraction of state index features include: Speed anomaly is determined based on flight speed parameters, attitude anomaly is determined based on flight attitude parameters, vibration anomaly is determined based on airframe vibration parameters, battery health is determined based on battery health parameters, and motor load is determined based on motor load parameters. The status index features are composed of speed anomaly, attitude anomaly, vibration anomaly, battery health and motor load.
[0009] In one possible design, the method further includes: When the speed anomaly exceeds the set speed anomaly threshold, the attitude anomaly exceeds the set attitude anomaly threshold, the vibration anomaly exceeds the set vibration anomaly threshold, the battery health exceeds the set battery health threshold, and / or the motor load exceeds the set motor load threshold, the system outputs a second anomaly warning message for the target UAV and sends a mission pause and avoidance command to the target UAV.
[0010] In one possible design, the flight trajectory evaluation based on flight position parameters and planned flight trajectory, and the extraction of trajectory index features, include: Based on the flight longitude parameters, flight latitude parameters, and flight altitude parameters, as well as the planned longitude data, planned latitude data, and planned altitude data corresponding to the four-dimensional flight trajectory, determine the nearest trajectory point on the planned flight trajectory corresponding to the current position of the target UAV, and the horizontal and vertical position deviations between the current position of the target UAV and the nearest trajectory point. The planning time corresponding to the nearest trajectory point is determined based on the planning time data corresponding to the four-dimensional flight trajectory, and the time deviation is calculated based on the planning time and the current time point; The trajectory index features are composed of horizontal position deviation, vertical position deviation, and time deviation.
[0011] In one possible design, the method further includes: The yaw spatial distance between the current position of the target UAV and the nearest trajectory point on the planned flight path is calculated using horizontal and vertical position deviations. Retrieve the yaw spatial distances of the target UAV relative to the planned flight trajectory at each historical time point before the current time point, and use the yaw spatial distances corresponding to the current time point and the yaw spatial distances corresponding to each historical time point to form a yaw spatial distance sequence. The yaw spatial distance sequence is input into a pre-set trend analysis model to perform yaw trend analysis, and the yaw trend analysis results of the target UAV are obtained. The trend analysis model adopts a pre-trained ARIMA model. When the yaw trend analysis results indicate that the target UAV's yaw trend is increasing, the system outputs a third abnormality warning for the target UAV and sends a trajectory adjustment control command to the target UAV.
[0012] Secondly, it provides a big data-based UAV management system, including a mission planning unit, flight control docking unit, mission monitoring unit, flight evaluation unit, feature fusion unit, joint detection unit, and early warning and control unit, among which: The mission planning unit is used to acquire the flight mission data of the target UAV and perform big data matching of UAV flight missions in the target airspace based on the flight mission data of the target UAV to determine the planned flight trajectory of the target UAV in the target airspace. The flight control docking unit is used to generate flight control data for the target UAV based on its planned flight trajectory in the target airspace, and send the flight control data to the target UAV so that the target UAV can perform flight missions in the target airspace according to the flight control data. The mission monitoring unit is used to collect the flight status parameters and flight position parameters of the target UAV in real time during the flight mission performed by the target UAV in the target airspace. The flight evaluation unit is used to evaluate flight status based on flight status parameters and extract status index features, and to evaluate flight trajectory based on flight position parameters and planned flight trajectory and extract trajectory index features. The feature fusion unit is used to fuse state index features and trajectory index features to obtain multi-index joint features; The joint detection unit is used to input the joint features of multiple indicators into a pre-set multi-indicator joint anomaly detection model to perform anomaly detection and obtain joint anomaly detection results. The early warning and control unit is used to output the first abnormality warning information of the target UAV when it is determined that the target UAV has a joint abnormality based on the joint anomaly detection results, and to send an emergency landing control command to the target UAV.
[0013] Thirdly, it provides a big data-based drone management system, including: Memory, used to store instructions; The processor is configured to read instructions stored in the memory and execute any one of the big data-based drone management methods described in the first aspect above, according to the instructions.
[0014] Fourthly, a computer-readable storage medium is provided, on which instructions are stored, which, when executed on a computer, cause the computer to perform any one of the big data-based drone management methods described in the first aspect. Simultaneously, a computer program product is also provided, which, when executed on a computer, performs any one of the big data-based drone management methods described in the first aspect.
[0015] Beneficial Effects: This invention acquires data on UAV flight missions, performs big data analysis, and plans flight trajectories to guide the UAVs to execute missions along these planned paths. During mission execution, real-time data collection of flight status and position parameters is used for joint feature analysis. When anomalies are detected, corresponding support and control measures are implemented to achieve efficient and reliable UAV flight management. Based on big data matching analysis, this invention enables efficient UAV trajectory planning. Through deep fusion analysis of multi-source heterogeneous data, reliable flight status perception is achieved, improving the timeliness and accuracy of flight anomaly identification. Differentiated control measures are implemented based on different flight anomaly determinations, balancing safety and efficiency to form a complete closed-loop UAV management system. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 This is a flowchart illustrating the method in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the system configuration in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the system configuration in Embodiment 3 of the present invention. Detailed Implementation
[0018] It should be noted that the descriptions of these embodiments are intended to aid in understanding the invention and do not constitute a limitation thereof. The specific structural and functional details disclosed herein are merely for describing exemplary embodiments of the invention. However, the invention may be embodied in many alternative forms and should not be construed as being limited to the embodiments described herein.
[0019] It should be understood that, unless otherwise explicitly specified and limited, the corresponding terms should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in the embodiments according to the specific circumstances.
[0020] Specific details are provided in the following description to provide a complete understanding of the exemplary embodiments. However, those skilled in the art will understand that the exemplary embodiments can be implemented without these specific details. For example, the system may be shown in block diagrams to avoid obscuring the example with unnecessary details. In other embodiments, well-known processes, structures, and techniques may be shown without non-essential details to avoid obscuring the embodiments.
[0021] Example 1: This embodiment provides a big data-based drone management method, which can be applied to corresponding drone management platforms. For example... Figure 1 As shown, the method includes the following steps: S1. Acquire the flight mission data of the target UAV, and perform big data matching of UAV flight missions in the target airspace based on the flight mission data of the target UAV to determine the planned flight trajectory of the target UAV in the target airspace.
[0022] In practice, the platform first acquires the flight mission data of the target UAV and determines its starting and ending waypoints within the target airspace based on this data. Then, it searches the existing big data of UAV flight missions in the target airspace for all historical flight trajectories matching the starting and ending waypoints. Based on this data, it identifies currently unoccupied historical flight trajectories and uses them as pre-selected flight trajectories. Finally, it selects the shortest flight trajectory from all pre-selected trajectories as the planned flight trajectory for the target UAV within the target airspace. This planned flight trajectory is a four-dimensional trajectory containing planned longitude, planned latitude, planned altitude, and planned time data.
[0023] S2. Generate flight control data for the target UAV based on its planned flight trajectory within the target airspace, and send the flight control data to the target UAV so that the target UAV can perform flight missions within the target airspace according to the flight control data.
[0024] In practice, the platform can generate flight control data for the target drone based on its planned flight trajectory within the target airspace, and then send the flight control data to the target drone so that the target drone can perform corresponding flight tasks within the target airspace based on the flight control data.
[0025] S3. During the flight mission of the target UAV within the target airspace, collect the flight status parameters and flight position parameters of the target UAV in real time.
[0026] In practice, the platform collects real-time flight status and position parameters of the target UAV while it is performing its flight mission within the target airspace. Flight status parameters include flight speed, flight attitude, airframe vibration, battery health, and motor load. Flight position parameters include longitude, latitude, and altitude.
[0027] S4. Evaluate the flight status based on flight status parameters and extract status index features. Evaluate the flight trajectory based on flight position parameters and planned flight trajectory and extract trajectory index features.
[0028] In practical implementation, the platform can determine speed anomaly based on flight speed parameters, such as dividing the flight speed parameter by the set reference speed or the planned speed contained in the flight mission data; determine attitude anomaly based on flight attitude parameters, such as calculating the deviation between the flight attitude parameters and normal flight attitude parameters; determine vibration anomaly based on airframe vibration parameters, such as using the root mean square value of high-frequency vibration acceleration as the vibration anomaly; determine battery health based on battery health parameters, such as determining the deviation of battery voltage or battery current from the standard voltage or current value and using the corresponding deviation as the battery health; and determine motor load based on motor load parameters, such as determining the motor load current based on the motor load parameters and using the ratio of the load current to the rated current as the motor load. Then, speed anomaly, attitude anomaly, vibration anomaly, battery health, and motor load are used to form a status indicator feature.
[0029] Simultaneously, based on flight longitude, latitude, and altitude parameters, as well as the planned longitude, latitude, and altitude data corresponding to the four-dimensional flight trajectory, the platform determines the nearest trajectory point on the planned flight trajectory corresponding to the target UAV's current position at the current time, and the horizontal and vertical positional deviations between the target UAV's current position and the nearest trajectory point. Then, based on the planned time data corresponding to the four-dimensional flight trajectory, the platform determines the planned time corresponding to the nearest trajectory point, and calculates the time deviation based on the planned time and the current time. Finally, the horizontal, vertical, and time deviations are used to form trajectory indicator features.
[0030] S5. Perform feature fusion on the state indicator features and trajectory indicator features to obtain multi-indicator joint features.
[0031] In practice, after obtaining the status indicator features and trajectory indicator features, the platform can fuse the two features to obtain multi-indicator joint features.
[0032] S6. Input the joint features of multiple indicators into a pre-set joint anomaly detection model to perform anomaly detection and obtain the joint anomaly detection results.
[0033] In practical implementation, the platform can input the joint features of multiple indicators into a pre-set multi-indicator joint anomaly detection model for multi-indicator joint anomaly detection, identifying correlational or latent anomalies that are difficult to detect through a single indicator, and obtaining the corresponding joint anomaly detection result, i.e., whether a joint anomaly exists. The multi-indicator joint anomaly detection model can be a pre-trained state-space model.
[0034] S7. When the target UAV is determined to have a joint anomaly based on the joint anomaly detection results, the first anomaly warning information of the target UAV is output, and an emergency landing control command is sent to the target UAV.
[0035] In practice, when the platform determines that the target drone has a joint abnormal situation based on the joint anomaly detection results, it directly outputs the first anomaly warning information of the target drone to prompt the management personnel to take countermeasures, and sends an emergency landing control command to the target drone to initiate the emergency landing procedure.
[0036] When the platform determines that the speed anomaly exceeds the set speed anomaly threshold, the attitude anomaly exceeds the set attitude anomaly threshold, the vibration anomaly exceeds the set vibration anomaly threshold, the battery health exceeds the set battery health threshold, and / or the motor load exceeds the set motor load threshold, it outputs a second anomaly warning message for the target UAV to prompt the management personnel to take countermeasures and sends a mission pause and avoidance command to the target UAV, causing the target UAV to pause the mission and perform standard avoidance actions, such as hovering, ascending, or returning to home.
[0037] Simultaneously, the platform can calculate the yaw spatial distance between the current position of the target UAV and the nearest trajectory point on the planned flight path using horizontal and vertical position deviations. It then retrieves the yaw spatial distances of the target UAV relative to the planned flight path from historical time points prior to the current time point, and uses these distances to form a yaw spatial distance sequence. This sequence is then input into a pre-set trend analysis model for yaw trend analysis, yielding the target UAV's yaw trend analysis results. The trend analysis model employs a pre-trained ARIMA model. When the platform determines that the target UAV's yaw trend is increasing based on the yaw trend analysis results, it outputs a third abnormality warning for the target UAV to prompt management personnel to take countermeasures and sends a trajectory adjustment control command to the target UAV, causing it to adjust its flight control parameters to conform to the planned flight path.
[0038] This method, based on big data matching analysis, enables efficient UAV trajectory planning. Through deep fusion analysis of multi-source heterogeneous data, it achieves reliable flight status perception, improves the timeliness and accuracy of flight anomaly identification, and implements differentiated control based on different flight anomaly determinations, balancing safety and efficiency to form a complete UAV management closed loop.
[0039] Example 2: This embodiment provides a drone management system based on big data, such as... Figure 2 As shown, it includes a mission planning unit, a flight control docking unit, a mission monitoring unit, a flight evaluation unit, a feature fusion unit, a joint detection unit, and an early warning and control unit, wherein: The mission planning unit is used to acquire the flight mission data of the target UAV and perform big data matching of UAV flight missions in the target airspace based on the flight mission data of the target UAV to determine the planned flight trajectory of the target UAV in the target airspace. The flight control docking unit is used to generate flight control data for the target UAV based on its planned flight trajectory in the target airspace, and send the flight control data to the target UAV so that the target UAV can perform flight missions in the target airspace according to the flight control data. The mission monitoring unit is used to collect the flight status parameters and flight position parameters of the target UAV in real time during the flight mission performed by the target UAV in the target airspace. The flight evaluation unit is used to evaluate flight status based on flight status parameters and extract status index features, and to evaluate flight trajectory based on flight position parameters and planned flight trajectory and extract trajectory index features. The feature fusion unit is used to fuse state index features and trajectory index features to obtain multi-index joint features; The joint detection unit is used to input the joint features of multiple indicators into a pre-set multi-indicator joint anomaly detection model to perform anomaly detection and obtain joint anomaly detection results. The early warning and control unit is used to output the first abnormality warning information of the target UAV when it is determined that the target UAV has a joint abnormality based on the joint anomaly detection results, and to send an emergency landing control command to the target UAV.
[0040] Example 3: This embodiment provides a drone management system based on big data, such as... Figure 3 As shown, at the hardware level, it includes: The data interface is used to establish data communication between the processor and external data terminals; Memory, used to store instructions; The processor is used to read instructions stored in the memory and execute the big data-based drone management method in Embodiment 1 according to the instructions.
[0041] Optionally, the system also includes an internal bus, through which the processor, memory, and data interface can be interconnected. This internal bus can be a PCIe (Peripheral Component Interconnect Eexpress) bus, which can be divided into an address bus, a data bus, a control bus, etc. The memory can include, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Flash Memory, First Input First Output (FIFO), and / or First In Last Out (FILO). The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0042] Example 4: This embodiment provides a computer-readable storage medium storing instructions. When these instructions are executed on a computer, the computer performs the big data-based drone management method described in Embodiment 1. The computer-readable storage medium refers to a data storage medium, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or Memory Sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems.
[0043] This embodiment also provides a computer program product that, when run on a computer, executes the big data-based drone management method described in Embodiment 1. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable system.
[0044] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for managing drones based on big data, characterized in that, The method comprises the following steps: acquiring flight task data of a target unmanned aerial vehicle (UAV), and performing UAV flight task big data matching on a target airspace based on the flight task data of the target UAV to determine a planned flight trajectory of the target UAV in the target airspace; generating flight control data of the target UAV based on the planned flight trajectory of the target UAV in the target airspace, and sending the flight control data to the target UAV to enable the target UAV to perform a flight task in the target airspace according to the flight control data; collecting flight state parameters and flight position parameters of the target UAV in real time during the performance of the flight task by the target UAV in the target airspace; performing flight state evaluation based on the flight state parameters to extract state index features, and performing flight trajectory evaluation based on the flight position parameters and the planned flight trajectory to extract trajectory index features; performing feature fusion on the state index features and the trajectory index features to obtain multi-index joint features; inputting the multi-index joint features into a preset multi-index joint anomaly detection model to perform anomaly detection, and obtaining joint anomaly detection results; when it is determined that the target UAV has a joint abnormal condition based on the joint anomaly detection results, outputting first abnormal early warning information of the target UAV, and sending an emergency landing control instruction to the target UAV. 2.The big data based UAV management method of claim 1, wherein, The method further comprises the following steps: determining a starting waypoint and an ending waypoint of the target UAV in the target airspace based on the flight task data of the target UAV; searching for all historical flight task trajectories from the starting waypoint to the ending waypoint from UAV flight task big data of the target airspace, and determining a currently unoccupied historical flight task trajectory based on the UAV flight task big data of the target airspace, wherein the currently unoccupied historical flight task trajectory is used as a preselected flight trajectory; selecting one of all preselected flight trajectories with the shortest flight time as the planned flight trajectory of the target UAV in the target airspace. 3.The big data based UAV management method of claim 1, wherein, The flight state parameters comprise flight speed parameters, flight attitude parameters, body vibration parameters, battery health parameters, and motor load parameters, the flight position parameters comprise flight longitude parameters, flight latitude parameters, and flight altitude parameters, and the planned flight trajectory is a four-dimensional flight trajectory comprising planned longitude data, planned latitude data, planned altitude data, and planned time data. 4.The big data based UAV management method of claim 3, wherein, The method further comprises the following steps: determining a speed anomaly degree based on the flight speed parameters, determining an attitude anomaly degree based on the flight attitude parameters, determining a vibration anomaly degree based on the body vibration parameters, determining a battery health degree based on the battery health parameters, and determining a motor load degree based on the motor load parameters; using the speed anomaly degree, the attitude anomaly degree, the vibration anomaly degree, the battery health degree, and the motor load degree to form the state index features. 5.The big data based UAV management method of claim 4, wherein, The method further comprises the following steps: When it is determined that the speed abnormality degree exceeds the set speed abnormality threshold, the attitude abnormality degree exceeds the set attitude abnormality threshold, the vibration abnormality degree exceeds the set vibration abnormality threshold, the battery health degree exceeds the set battery health threshold, and / or the motor load degree exceeds the set motor load threshold, second abnormal early warning information of the target UAV is output, and a task suspension risk avoidance instruction is sent to the target UAV. 6.The big data based UAV management method of claim 3, wherein, The flight trajectory evaluation based on the flight position parameters and the planned flight trajectory extracts trajectory index features, including: According to the flight longitude parameter, the flight latitude parameter, the flight height parameter, the planned longitude data, the planned latitude data, and the planned height data corresponding to the four-dimensional flight trajectory, the nearest trajectory point corresponding to the position of the target UAV at the current time point on the planned flight trajectory is determined, as well as the horizontal position deviation and the vertical position deviation between the position of the target UAV at the current time point and the nearest trajectory point; The planned time corresponding to the nearest trajectory point is determined according to the planned time data corresponding to the four-dimensional flight trajectory, and the time deviation is calculated based on the planned time and the current time point; The horizontal position deviation, the vertical position deviation, and the time deviation are used to form the trajectory index features. 7.The big data based UAV management method of claim 6, wherein, The method further includes: The horizontal position deviation and the vertical position deviation are used to calculate the off-course distance between the position of the target UAV at the current time point and the nearest trajectory point on the planned flight trajectory; The off-course distance of the target UAV at each historical time point relative to the planned flight trajectory before the current time point is retrieved, and the off-course distance corresponding to the current time point and the off-course distances corresponding to each historical time point are used to form an off-course distance sequence; The off-course distance sequence is input into a pre-set trend analysis model for off-course trend analysis, and the off-course trend analysis result of the target UAV is obtained, wherein the trend analysis model uses a pre-trained ARIMA model; When it is determined that the off-course trend of the target UAV increases according to the off-course trend analysis result, third abnormal early warning information of the target UAV is output, and a flight path adjustment control instruction is sent to the target UAV.
8. A drone management system based on big data, characterized by, The system includes a task planning unit, a flight control interfacing unit, a task monitoring unit, a flight evaluation unit, a feature fusion unit, a joint detection unit, and an early warning control unit, wherein: The task planning unit is configured to obtain flight task data of the target UAV, and perform UAV flight task big data matching in the target airspace based on the flight task data of the target UAV to determine a planned flight trajectory of the target UAV in the target airspace; The flight control interfacing unit is configured to generate flight control data of the target UAV based on the planned flight trajectory of the target UAV in the target airspace, and send the flight control data to the target UAV to enable the target UAV to perform a flight task in the target airspace according to the flight control data; The task monitoring unit is configured to collect flight state parameters and flight position parameters of the target UAV in real time during the performance of the flight task by the target UAV in the target airspace; The flight evaluation unit is configured to perform flight state evaluation based on the flight state parameters to extract state index features, and perform flight trajectory evaluation based on the flight position parameters and the planned flight trajectory to extract trajectory index features; The feature fusion unit is configured to perform feature fusion on the state indicator features and the trajectory indicator features to obtain multi-indicator joint features. The joint detection unit is configured to input the multi-indicator joint features into a preset multi-indicator joint anomaly detection model to perform anomaly detection and obtain joint anomaly detection results. The early warning management unit is configured to output first anomaly early warning information of the target UAV and send an emergency landing control instruction to the target UAV when it is determined that the target UAV has a joint abnormal condition according to the joint anomaly detection results.
9. A drone management system based on big data, characterized by, The computer program product comprises: a memory configured to store instructions; a processor configured to read the instructions stored in the memory and execute the big data-based UAV management method according to any one of claims 1-7.
10. A computer program product, characterised in that, When the computer program product runs on the computer, the big data-based UAV management method according to any one of claims 1-7 is executed.