A multi-uav dynamic cooperative positioning method for low-altitude airspace moving targets

Abstract

The application discloses a kind of low air space mobile target's multi-unmanned plane dynamic cooperative positioning method, method includes: planning N carries sensor system's unmanned plane in airport protection area and utilizes each sensor system jointly to gather mobile target's joint monitoring data jointly by data link transmission to target monitoring system;Target monitoring system utilizes joint monitoring data to calculate the maneuvering intensity factor of mobile target Dynamic adjustment sampling period and dynamically select interpolation method;S3, based on the joint monitoring data after sampling interpolation, the position solution of mobile target is carried out.The application can dynamically adjust sampling period and dynamically select interpolation method to carry out the sampling interpolation processing of joint monitoring data, simultaneously, unmanned plane will be controlled to approach mobile target to facilitate more accurate data acquisition, finally, the data that satisfies the requirement of positioning accuracy is obtained, and the accurate positioning solution of mobile target is carried out through target position solution matrix, which significantly improves the accuracy and real-time performance of cooperative positioning.

Description

Technical Field

[0001] This invention relates to the field of dynamic adjustment and cooperative positioning of multiple UAV swarms, and in particular to a method for dynamic cooperative positioning of multiple UAVs targeting moving targets in low-altitude airspace, providing a technical foundation for low-altitude safety. Background Technology

[0002] Against the backdrop of the booming low-altitude economy, the perception and collaborative positioning of drones within low-altitude integrated airspace are becoming increasingly important for aircraft flight safety. This is especially true for non-cooperative drones in low-altitude airfields, which are often not effectively regulated and protected under the existing airspace management system. Therefore, perceiving and identifying such drones, and accurately locating and coordinating their dynamic positioning, can effectively fill existing technological gaps in the safety field, providing an important technological foundation for the safety management of the low-altitude economy, and offering new technological paths and approaches for low-altitude airspace management.

[0003] Airport security protection zones are crucial for airport operations. These zones include the airport's clear airspace (which must prevent moving targets from entering the airspace and is a vital protection area) and the airport's near-ground protection zone (including the airport ground area, where many airport facilities are located, such as control towers, runways, meteorological observation equipment, navigation and communication equipment, etc.). These facilities also need to be protected against moving targets, as falling or colliding with them can damage them. When a moving target approaches or enters the airport security protection zone, the airport first attempts to drive it away. If this is not possible, the airport then monitors, locates, intercepts, or destroys the target. Moving targets include birds, hot air balloons, and drones. Drones, in particular, are significant. With the continuous improvement of drone performance, their maneuverability and speed have increased rapidly. If a moving target is highly maneuverable, its high speed and maneuverability cause its position to change rapidly, requiring drones to adjust their strategies for positioning. In existing technologies, when facing moving targets with varying degrees of maneuverability, UAVs cannot dynamically adjust their data sampling and positioning strategies based on the target's maneuverability. Consequently, when facing highly maneuverable and flexible moving targets, they often suffer from low positioning accuracy. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-UAV dynamic cooperative positioning method for moving targets in low-altitude airspace. This method utilizes multiple UAVs to monitor the moving target and obtain joint monitoring data. Based on the judgment results, the sampling period is dynamically adjusted to sample the joint monitoring data, and a dynamic interpolation method is selected to interpolate the data between the sampled data. At the same time, some or all UAVs are controlled to approach the moving target to facilitate more accurate data acquisition. Finally, the sampled and interpolated joint monitoring data that meets the positioning accuracy requirements is obtained, and the precise positioning of the moving target is calculated using a target position calculation matrix.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A method for dynamic cooperative localization of multiple UAVs targeting moving targets in low-altitude airspace, the method comprising:

[0007] S1. Plan N drones equipped with sensor systems to fly collaboratively within the airport protection area and use the various sensor systems to jointly collect joint monitoring data of moving targets and transmit it to the target monitoring system via data link;

[0008] S2. The target monitoring system uses joint monitoring data to calculate the maneuver intensity factor of the moving target based on the maneuver performance index, sets the maneuver intensity threshold, and if the maneuver intensity factor is less than or equal to the maneuver intensity threshold, it samples data from each sensor system according to the preset sampling period T1; if the maneuver intensity factor is greater than the maneuver intensity threshold, it dynamically reduces the sampling period T1 and controls some or all UAVs to approach the moving target.

[0009] S3. The target monitoring system samples joint monitoring data in real time and constructs a target position calculation matrix to calculate the position of the moving target.

[0010] To better implement this invention, in method S1, the sensor system includes several airborne sensors and / or an optoelectronic tracking system; the joint monitoring data includes UAV position data, target relative angle data, and target maneuvering attitude data. The target relative angle data consists of the target relative azimuth angle data and pitch angle data of the moving target collected by each UAV, and the target relative azimuth angle data and pitch angle data are the angle data recorded by the sensor system when the UAV detects the moving target; the target maneuvering attitude data is the angular velocity of the moving target at time t. Linear acceleration Turning radius accelerometer .

[0011] Preferably, in method S2, the target monitoring system obtains the maneuver intensity factor of the moving target at time t. The calculation expression is as follows:

[0012] ,in , , , These are the set maneuver thresholds corresponding to angular velocity, linear acceleration, turning radius, and jerk, respectively. , , , These correspond to the weight parameters.

[0013] This invention provides a first method for reducing the sampling period when the maneuver intensity factor is greater than the maneuver intensity threshold, as follows: In method S2, if the maneuver intensity factor at time t is greater than the maneuver intensity threshold... ,by Data is sampled from each sensor system using the reduced sampling period.

[0014] This invention provides a second method for reducing the sampling period when the maneuver intensity factor is greater than the maneuver intensity threshold, as follows: In method S2, if the maneuver intensity factor at time t is greater than the maneuver intensity threshold... Several maneuver intensity intervals are defined according to the maneuver intensity factor. Each maneuver intensity interval corresponds to a sampling period or a sampling period reduction rate. The maneuver intensity factor at time t is then determined. The sampling period is dynamically reduced and adjusted according to the motor intensity range to which it belongs.

[0015] Preferably, in method S2, if the maneuver intensity factor is greater than the maneuver intensity threshold, the distance between the UAV and the moving target is calculated and sorted from farthest to closest. At the same time, flight control is performed on some or all UAVs in the order from farthest to closest to approach the moving target.

[0016] Preferably, in method S3, when the target monitoring system samples joint monitoring data, the following dynamic interpolation adjustment method is used:

[0017] The detection sampling interval of airborne sensors or photoelectric tracking systems in sensor systems Compare with the sampling period corresponding to time t; if , If the sampling period corresponding to time t after dynamic adjustment is given, then linear interpolation is used for data interpolation; otherwise, higher-order interpolation is used, such as higher-order polynomial interpolation, piecewise higher-order polynomial interpolation, Lagrange higher-order interpolation, or Hermite higher-order interpolation.

[0018] Preferably, the target monitoring system samples joint monitoring data at the study time and extracts UAV position data and target relative angle data, and uses the least squares method to calculate the position vector of the moving target at the study time. The expression is as follows: ,in The coefficient matrix is ​​composed of UAV angle data. This is a constant vector composed of the UAV's position and angle data. Coefficient matrix The transpose of the matrix, for The inverse of a matrix.

[0019] Preferably, the present invention further includes the following method:

[0020] S4. Following the position calculation method at the research time in method S3, calculate the position of the moving target sequentially along the time axis and construct the temporal position data of the moving target. While acquiring the temporal position data, continuously monitor and select suitable attitude opportunities for command and interception. Construct a moving target time series prediction model, which includes an ARIMA model or a Transformer model. The moving target time series prediction model uses the temporal position data to predict the position of the moving target at future times.

[0021] Preferably, in method S1, a ground monitoring station is constructed, and the target monitoring system is installed in the ground monitoring station and / or a certain UAV; the airport protection area includes the airport airspace protection zone and / or the airport near-ground protection zone.

[0022] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0023] (1) This invention utilizes multiple UAVs to monitor moving targets and obtain joint monitoring data. First, the maneuver intensity factor of the moving target is calculated and the maneuver intensity threshold is compared and judged. Based on the judgment result, the sampling period is dynamically adjusted to sample the joint monitoring data and a dynamic interpolation method is selected to interpolate the data between the sampled data. At the same time, some or all UAVs are controlled to approach the moving target to obtain data more accurately. Finally, the sampling interpolated joint monitoring data that meets the positioning accuracy requirements is obtained. The target position solution matrix is ​​used to accurately locate the moving target and obtain high-precision moving target position data and position time series data.

[0024] (2) By introducing a maneuver intensity factor and dynamically dividing and setting different sampling periods and interpolation methods accordingly, this invention is suitable for dynamic processing of different maneuver intensity factors. It can adaptively and flexibly adjust the positioning strategy according to the moving target to improve data density and positioning accuracy. In particular, for high-speed and high-maneuver targets, it significantly improves the accuracy and real-time performance of cooperative positioning.

[0025] (3) This invention comprehensively considers multiple key maneuver performance indicators such as angular velocity and linear acceleration, which can more comprehensively and accurately reflect the instantaneous maneuver state of the moving target, making the selection of interpolation method more in line with the actual scenario requirements; the interpolation method is dynamically selected according to the maneuver intensity, taking into account both computational efficiency and positioning accuracy, providing reliable technical support for the defense positioning of clustered UAVs, and improving the speed and robustness requirements of positioning task completion. Attached Figure Description

[0026] Figure 1 This is a flowchart of the multi-UAV dynamic cooperative positioning method of the present invention;

[0027] Figure 2 This is a spatial diagram illustrating the dynamic collaborative positioning of two drones to a moving target, as illustrated in Example 1. Detailed Implementation

[0028] The present invention will be further described in detail below with reference to embodiments:

[0029] Example 1

[0030] like Figure 1 As shown, a multi-UAV dynamic cooperative localization method for moving targets in low-altitude airspace is described, the method comprising:

[0031] S1. Plan N (N≥2) drones equipped with sensor systems to fly collaboratively within the airport protection zone. Utilizing various sensor systems (preferably, the sensor system includes several airborne sensors and / or an optoelectronic tracking system; airborne sensors include radio frequency sensors (RF sensors), radar sensors, visual sensors, and infrared sensors, etc.), they jointly collect joint monitoring data of moving targets and transmit it to the target monitoring system via a data link. The airport protection zone includes the airport's clear airspace and / or near-ground protection zone. In this embodiment, a ground monitoring station is constructed, and the target monitoring system is located at the ground monitoring station and / or on one of the drones. All joint monitoring data is unified to the same coordinate system, and the data from different drones are time-registered to the same time system. The joint monitoring data is standard data within the same coordinate system and time system. Preferably, the present invention first identifies the moving target through the sensor system on the drone. The moving target is identified as a drone, a bird, a hot air balloon, etc. If it is a drone, it is determined whether it is a cooperative drone. If it is a non-cooperative drone, N drones equipped with sensor systems fly together to track and monitor the non-cooperative drone. If it is a cooperative drone, the N drones equipped with sensor systems do not take any action.

[0032] In some embodiments, the joint monitoring data includes UAV position data, target relative angle data, and target maneuvering attitude data, wherein the UAV position data is the location data of the UAV positioning, and the position data corresponding to UAV i is... The target relative angle data consists of the target relative azimuth and pitch angle data collected by each UAV from the moving target. , Here are the coordinates of drone i. To detect the relative azimuth angle of a moving target for UAV i. The pitch angle is used to detect the moving target for UAV i. The target relative azimuth and pitch angle data are the angle data recorded by the sensor system during the UAV's detection of the moving target; the target maneuver attitude data is the angular velocity of the moving target at time t. Linear acceleration Turning radius accelerometer .

[0033] S2. The target monitoring system uses joint monitoring data to calculate the maneuver intensity factor of the moving target based on maneuver performance indicators, and sets a maneuver intensity threshold (this invention uses a maneuver intensity threshold). (This involves quantifying the maneuverability of a moving target at time t). In some embodiments, the target monitoring system obtains the maneuverability factor of the moving target at time t. The calculation expression is as follows:

[0034] ,in , , , These are the set maneuver thresholds corresponding to angular velocity, linear acceleration, turning radius, and jerk, respectively. , , , These correspond to the weight parameters, , , , The range of values ​​is And satisfy .

[0035] If the mobility intensity factor is less than or equal to the mobility intensity threshold (If the moving target has weak maneuverability, i.e., the target moves slowly), then data is sampled from each sensor system according to the preset sampling period T1; if the maneuver intensity factor is greater than the maneuver intensity threshold (indicating that the moving target has strong maneuverability, i.e., the target moves at high speed), then the sampling period T1 is dynamically reduced and some or all UAVs are controlled to approach the moving target.

[0036] This invention provides a first method for reducing the sampling period when the maneuver intensity factor is greater than the maneuver intensity threshold, as follows: If the maneuver intensity factor at time t (i.e., the study time) is greater than the maneuver intensity threshold... ,by Data is sampled from each sensor system using the reduced sampling period.

[0037] S3. The target monitoring system samples joint monitoring data in real time and constructs a target position calculation matrix to calculate the position of the moving target. When the target monitoring system samples joint monitoring data, the following dynamic interpolation adjustment method is used:

[0038] The detection sampling interval of airborne sensors or photoelectric tracking systems in sensor systems Compare with the sampling period corresponding to time t; if , If the sampling period corresponding to time t after dynamic adjustment is given, then linear interpolation is used for data interpolation; otherwise, higher-order interpolation methods are used, such as high-order polynomial interpolation, piecewise high-order polynomial interpolation, Lagrange high-order interpolation, or Hermitian high-order interpolation. After the target monitoring system samples the joint monitoring data in real time, the interpolation process is then performed using the selected interpolation method described above to obtain the position data of all UAVs at time t (time t is stored in timestamp form along with position and angle data, forming time-series data of position and target relative angle). Relative angle data to the target .

[0039] In some embodiments, the target monitoring system samples joint monitoring data at the time of study and extracts UAV location data. Relative angle data to the target , The coordinate data of UAV i at the time of study. To enable UAV i to detect the relative azimuth angle of a moving target at the time of study, The pitch angle of the moving target detected by UAV i at the study time is calculated. The UAV position data, the target relative angle data, and the position of the moving target to be solved are then compared. The following system of equations can be constructed jointly: Linearizing the system of equations yields the following matrix: ;

[0040] , ;in Let be the position vector of the moving target at the time of study to be solved. The coefficient matrix, Let be a constant vector consisting of the position and angle of the UAV.

[0041] This embodiment uses the least squares method to obtain the position vector of the moving target at the time of study. The expression is as follows: ,in Coefficient matrix The transpose of the matrix, for The inverse of a matrix.

[0042] like Figure 2 As shown in the example, this embodiment uses two drones equipped with sensor systems, namely drone A and drone B, and the moving target is the incoming drone. Figure 2 The moving target shown is located in UAV A, and the target monitoring system is placed in UAV A. The airport protection zone is selected as a facility within the airport's near-ground protection zone (both its near-ground area and the airport's airspace protection zone belong to the airport protection zone). Both UAVs are unified to the same standard coordinate system (such as the Northeast coordinate system or others). In this embodiment... Figure 2 Taking a large coordinate system centered at a certain point as an example, where the X-axis points due east, the Y-axis points due north, and the Z-axis points vertically to the zenith, the azimuth angle of UAV i detecting a moving target is... See Figure 2 The pitch angle of UAV i for detecting moving targets See Figure 2 The multi-UAV dynamic cooperative localization method of the present invention, according to Embodiment 1, can calculate the position data of the moving target at the time of study.

[0043] Example 2

[0044] A method for dynamic cooperative localization of multiple UAVs targeting moving targets in low-altitude airspace, such as Figure 1 As shown, the method includes:

[0045] S1. Plan N (N≥2) drones equipped with sensor systems to fly collaboratively within the airport protection zone and jointly collect moving target monitoring data using their respective sensor systems (preferably, the sensor system includes several airborne sensors and / or an electro-optical tracking system), which is then transmitted to the target monitoring system via a data link. The airport protection zone includes the airport's clear airspace and / or near-ground protection zone. In this embodiment, a ground monitoring station is constructed, and the target monitoring system is located at the ground monitoring station and / or on one of the drones.

[0046] In some embodiments, the joint monitoring data includes UAV position data, target relative angle data, and target maneuvering attitude data, wherein the UAV position data is the location data of the UAV positioning, and the position data corresponding to UAV i is... The target relative angle data consists of the target relative azimuth and pitch angle data collected by each UAV from the moving target. , Here are the coordinates of drone i. To detect the relative azimuth angle of a moving target for UAV i. The pitch angle is used to detect the moving target for UAV i. The target relative azimuth and pitch angle data are the angle data recorded by the sensor system during the UAV's detection of the moving target; the target maneuver attitude data is the angular velocity of the moving target at time t. Linear acceleration Turning radius accelerometer .

[0047] S2. The target monitoring system calculates the maneuver intensity factor of the moving target based on maneuver performance indicators using joint monitoring data, sets a maneuver intensity threshold, and if the maneuver intensity factor is less than or equal to the threshold, it samples data from each sensor system according to a preset sampling period T1; if the maneuver intensity factor is greater than the threshold, it dynamically reduces the sampling period T1 and controls some or all of the UAVs to approach the moving target. In some embodiments, the target monitoring system obtains the maneuver intensity factor of the moving target at time t. The calculation expression is as follows:

[0048] ,in , , , These are the set maneuver thresholds corresponding to angular velocity, linear acceleration, turning radius, and jerk, respectively. , , , These correspond to the weight parameters.

[0049] This invention provides a second method for reducing the sampling period when the maneuver intensity factor is greater than the maneuver intensity threshold, as follows: If the maneuver intensity factor at time t is greater than the maneuver intensity threshold... Several maneuver intensity intervals are defined according to the maneuver intensity factor. Each maneuver intensity interval corresponds to a sampling period or a sampling period reduction rate. The maneuver intensity factor at time t is then determined. The sampling period is dynamically reduced and adjusted according to the maneuver intensity range to which the UAV belongs. If the maneuver intensity factor is greater than the maneuver intensity threshold, the distance between the UAV and the moving target is calculated and sorted from farthest to closest. At the same time, some or all UAVs are controlled to approach the moving target in the order from farthest to closest.

[0050] S3. The target monitoring system samples joint monitoring data in real time and constructs a target position calculation matrix to calculate the position of the moving target. When the target monitoring system samples joint monitoring data, the following dynamic interpolation adjustment method is used:

[0051] The detection sampling interval of airborne sensors or photoelectric tracking systems in sensor systems Compare with the sampling period corresponding to time t; if , If the sampling period corresponding to time t after dynamic adjustment is given, then linear interpolation is used for data interpolation; otherwise, higher-order interpolation is used, such as higher-order polynomial interpolation, piecewise higher-order polynomial interpolation, Lagrange higher-order interpolation, or Hermite higher-order interpolation.

[0052] In some embodiments, the target monitoring system samples joint monitoring data at the time of study and extracts UAV location data. Relative angle data to the target , The coordinate data of UAV i at the time of study. To enable UAV i to detect the relative azimuth angle of a moving target at the time of study, The pitch angle of the moving target detected by UAV i at the study time is calculated. The UAV position data, the target relative angle data, and the position of the moving target to be solved are then compared. The following system of equations can be constructed jointly: Linearizing the system of equations yields the following matrix: ;

[0053] , ;in Let be the position vector of the moving target at the time of study to be solved. The coefficient matrix, Let be a constant vector consisting of the position and angle of the UAV.

[0054] This embodiment uses the least squares method to obtain the position vector of the moving target at the time of study. The expression is as follows: ,in Coefficient matrix The transpose of the matrix, for The inverse of a matrix.

[0055] S4. Following the position calculation method at the research time in method S3, the position of the moving target is calculated sequentially along the time axis to form the temporal position data of the moving target. While acquiring the temporal position data, the system continuously monitors and selects suitable attitude timings for command and interception. In some embodiments, the present invention constructs a moving target time series prediction model, which includes an ARIMA model or a Transformer model. The moving target time series prediction model uses the temporal position data to predict the position of the moving target at future times. The temporal position data is time-series position data, and attitude data (including linear acceleration, angular velocity, jerk, etc.) is also time-series data. The moving target time series prediction model predicts the future arrival position. Alternatively, inertial inference can be performed based on the current position and attitude data (including linear acceleration, angular velocity, jerk, etc.) of the moving target, facilitating the prediction of the future position data of the moving target at future times.

[0056] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for dynamic cooperative localization of multiple unmanned aerial vehicles (UAVs) targeting moving targets in low-altitude airspace, characterized in that: The methods include: S1. Plan N drones equipped with sensor systems to fly collaboratively within the airport protection area and use the various sensor systems to jointly collect joint monitoring data of moving targets and transmit it to the target monitoring system via data link; S2. The target monitoring system uses joint monitoring data to calculate the maneuver intensity factor of the moving target based on maneuver performance indicators. The target monitoring system obtains the maneuver intensity factor of the moving target at time t. The calculation expression is as follows: ,in , , , These are the set maneuver thresholds corresponding to angular velocity, linear acceleration, turning radius, and jerk, respectively. , , , These correspond to weight parameters respectively; set a maneuver intensity threshold, and if the maneuver intensity factor is less than or equal to the maneuver intensity threshold, then sample data from each sensor system according to the preset sampling period T1; If the maneuver intensity factor is greater than the maneuver intensity threshold, the sampling period T1 is dynamically reduced and some or all UAVs are controlled to approach the moving target; if the maneuver intensity factor at time t is greater than the maneuver intensity threshold... ,by Data is sampled from each sensor system using the reduced sampling period; S3. The target monitoring system samples joint monitoring data in real time and constructs a target position calculation matrix to calculate the position of the moving target.

2. The multi-UAV dynamic cooperative positioning method for a moving target in low-altitude airspace according to claim 1, characterized in that: In method S1, the sensor system includes several airborne sensors and / or an optoelectronic tracking system; the joint monitoring data includes UAV position data, target relative angle data, and target maneuvering attitude data. The target relative angle data consists of the target relative azimuth and pitch angle data of the moving target acquired by each UAV, which are recorded by the sensor system when the UAV detects the moving target. The target maneuvering attitude data is the angular velocity of the moving target at time t. Linear acceleration Turning radius accelerometer .

3. A multi-UAV dynamic cooperative positioning method for a moving target in low-altitude airspace according to claim 1, characterized in that: In method S2, if the maneuver intensity factor at time t is greater than the maneuver intensity threshold... Several maneuver intensity intervals are defined according to the maneuver intensity factor. Each maneuver intensity interval corresponds to a sampling period or a sampling period reduction rate. The maneuver intensity factor at time t is then determined. The sampling period is dynamically reduced and adjusted according to the motor intensity range to which it belongs.

4. A multi-UAV dynamic cooperative positioning method for a moving target in low-altitude airspace according to claim 1, characterized in that: In method S2, if the maneuver intensity factor is greater than the maneuver intensity threshold, the distance between the UAV and the moving target is calculated and sorted from farthest to closest. At the same time, some or all UAVs are controlled to approach the moving target in the order from farthest to closest.

5. A multi-UAV dynamic cooperative positioning method for a moving target in low-altitude airspace according to claim 2, characterized in that: In method S3, when the target monitoring system samples joint monitoring data, the following dynamic interpolation adjustment method is used: The detection sampling interval of airborne sensors or photoelectric tracking systems in sensor systems Compare with the sampling period corresponding to time t; if , If the sampling period corresponding to time t after dynamic adjustment is given, then linear interpolation is used for data interpolation; otherwise, higher-order interpolation is used, such as higher-order polynomial interpolation, piecewise higher-order polynomial interpolation, Lagrange higher-order interpolation, or Hermite higher-order interpolation.

6. A multi-UAV dynamic cooperative positioning method for a moving target in low-altitude airspace according to claim 2, characterized in that: The target monitoring system samples joint monitoring data at the study time and extracts UAV position data and target relative angle data. The least squares method is used to calculate the position vector of the moving target at the study time. The expression is as follows: ,in The coefficient matrix is ​​composed of UAV angle data. This is a constant vector composed of the UAV's position and angle data. Coefficient matrix The transpose of the matrix, for The inverse of a matrix.

7. A method for dynamic cooperative localization of multiple UAVs targeting moving targets in low-altitude airspace according to claim 1, characterized in that: It also includes the following methods: S4. Following the position calculation method at the research time in method S3, calculate the position of the moving target sequentially along the time axis and construct the temporal position data of the moving target. While acquiring the temporal position data, continuously monitor and select suitable attitude opportunities for command and interception. Construct a moving target time series prediction model, which includes an ARIMA model or a Transformer model. The moving target time series prediction model uses the temporal position data to predict the position of the moving target at future times.

8. A multi-UAV dynamic cooperative positioning method for a moving target in low-altitude airspace according to claim 1, characterized in that: In method S1, a ground monitoring station is constructed, and the target monitoring system is set in the ground monitoring station or / and a certain UAV; the airport protection area includes the airport airspace protection zone or / and the airport near-ground protection zone.

Images