An unmanned aerial vehicle trajectory tracking method, server, medium and product

Abstract

The application provides a UAV trajectory tracking method, a server, a medium and a product, and relates to the technical field of UAV tracking. The method comprises the following steps: receiving GPS positioning signals of all working UAVs in real time; when it is monitored that the GPS positioning signal of any target working UAV is lost, determining a reference UAV of the target working UAV, and acquiring real-time positioning information of the reference UAV; calculating the historical relative position relationship between the reference UAV and the target working UAV; based on the historical relative position relationship and the real-time positioning information of the reference UAV, calculating the real-time virtual positioning of the target working UAV, and generating a temporary tracking trajectory according to the real-time virtual positioning; and issuing the temporary tracking trajectory to the target working UAV. The method can guarantee the continuity of the UAV trajectory and the integrity of the work, effectively avoid the yaw, collision or work interruption caused by the loss of position, and significantly improve the safety and reliability of the UAV cluster work in a complex scene.

Description

Technical Field

[0001] This application relates to the field of drone tracking technology, and in particular to a drone trajectory tracking method, server, medium, and product. Background Technology

[0002] With the rapid development of drone technology, its application scenarios have expanded from simple aerial photography to complex scenarios such as high-density swarm operations (e.g., drone fireworks displays and formation performances), large-scale agricultural plant protection, and power line inspection. In these scenarios, precise trajectory control of drones is crucial for ensuring operational safety, efficiency, and effectiveness, and the foundation of trajectory control relies on stable and high-precision positioning information.

[0003] Currently, the mainstream positioning method for drones is GPS positioning. The drone flight control system receives GPS positioning data in real time, compares it with the preset overall operation path (such as the flight path for agricultural plant protection or the timing coordinates of formation performance), calculates the deviation between the current position and the target path, and then corrects the flight attitude by adjusting the motor speed, servo angle, etc., to ensure that the drone flies along the preset trajectory.

[0004] However, in complex real-world operational scenarios, GPS positioning signals are susceptible to interference from various factors, leading to loss or drift. These include signal obstruction in densely built-up urban areas, electromagnetic interference near high-voltage lines, mutual interference between signals from multiple drones in swarm operations, and the attenuation of satellite signals by smoke during fireworks displays. When the GPS positioning signal of the target drone is lost, failure to obtain its accurate location in a timely manner can easily cause the drone to deviate from its preset trajectory, resulting in serious problems such as swarm collisions, missed work areas, or even equipment crashes. Summary of the Invention

[0005] This application provides a drone trajectory tracking method, server, medium, and product for supplementing the drone's flight trajectory data during periods of lost positioning signals, thereby ensuring mission continuity and flight safety.

[0006] In a first aspect, this application provides a method for tracking the trajectory of a drone, applied to a server. The method includes: receiving GPS positioning signals from all operational drones in real time; when the GPS positioning signal of any target operational drone is detected to be lost, determining a reference drone for the target operational drone and obtaining the real-time positioning information of the reference drone; calculating the historical relative position relationship between the reference drone and the target operational drone, the historical relative position relationship including horizontal relative distance, altitude difference, and relative azimuth angle; based on the historical relative position relationship and the real-time positioning information of the reference drone, calculating the real-time virtual positioning of the target operational drone, and generating a temporary tracking trajectory based on the real-time virtual positioning; and sending the temporary tracking trajectory to the target operational drone.

[0007] By employing the aforementioned technical solution, a mobile and reliable reference coordinate system is established for the target drone that has lost its signal by identifying a "reference drone" with a normal GPS signal. Secondly, the core lies in calculating and utilizing the "historical relative position relationship," which (including horizontal distance, altitude difference, and azimuth angle) is essentially a digital representation of the drone formation. By overlaying this relatively stable formation relationship onto the real-time precise position of the "reference drone," the current virtual position of the target drone can be realistically reproduced. The temporary trajectory generated based on this virtual positioning allows the target drone to continue following the formation and maintaining its formation even without its own GPS, thus greatly ensuring the continuity of the trajectory and the integrity of the operation. This effectively avoids deviations, collisions, or operational interruptions caused by position loss, significantly improving the safety and reliability of drone swarm operations in complex scenarios.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, the step of determining a reference drone when the GPS positioning signal of any target drone is detected to be lost includes: when the GPS positioning signal of any target drone is detected to be lost, obtaining the total operation path, the timestamp of signal loss, and the GPS positioning of the last valid loss point at the time of signal loss; based on the GPS positioning of the loss point, searching outwards according to a preset distance gradient to filter out candidate drones with valid GPS positioning signals reaching a set number of candidates; based on the total operation path and the timestamp, calculating the spatial matching degree between each candidate drone and the target drone; filtering out candidate drones with a spatial matching degree greater than a set spatial ratio, and calculating the straight-line distance between each candidate drone and the target drone at the time of GPS signal loss, selecting the drone with the shortest straight-line distance as the reference drone.

[0009] By employing the aforementioned technical solution, physically nearby UAVs are quickly located through a "search outwards in a predetermined distance gradient," forming a candidate pool. A key step involves introducing a forward-looking evaluation metric: "spatial matching degree." This not only considers the current distance but also compares the "expected path segments" over a future period to determine whether the candidate UAV and the target UAV are coordinated in their movement trends. A candidate with a high spatial matching degree means it will maintain a close escort relationship with the target UAV in subsequent missions. Finally, among all candidates with high path coordination, the one with the shortest straight-line distance is selected, ensuring the highest accuracy of the initial reference. This two-step screening method—"look at the trend first, then the distance"—avoids selecting a "neighbor" that is about to part ways as a reference, thus ensuring the long-term stability and reliability of the selected reference UAV and laying a solid foundation for the continuous accuracy of subsequent virtual positioning calculations.

[0010] In some embodiments of the first aspect, the step of calculating the spatial matching degree between each candidate drone and the target drone based on the total operation path and the timestamp includes: extracting the path location point of the target drone at the timestamp from the total operation path; taking the path location point as the starting point, extracting the expected path segment that the target drone should execute within a set continuation time; obtaining the planned path segment of each candidate drone within the set continuation time based on the timestamp; discretizing the expected path segment and each planned path segment into multiple time point coordinates at a preset time interval; calculating the spatial distance between the planned path segment and each expected path segment at each corresponding time point, and counting the number of time points where the spatial distance is less than a preset spatial threshold; and determining the spatial matching degree as the ratio of the number of time points to the total number of time points within the set continuation time.

[0011] By employing the aforementioned technical solution and statistically analyzing the percentage of time points within a future period where the spatial distance between the two drones is less than a "preset spatial threshold," this method can accurately quantify the degree of cooperative movement between the two drones while performing their respective tasks. A higher ratio indicates that they will be in close flight together for most of the time. This quantitative evaluation method eliminates subjective judgment in the selection of the reference drone, making it objective and precise. It ensures that the selected reference drone is kinematically highly correlated with the target drone, thus providing high-quality input data for the entire trajectory tracking scheme and improving the scientific rigor and accuracy of reference selection.

[0012] In some embodiments of the first aspect, the step of calculating the real-time virtual positioning of the target drone based on the historical relative positional relationship and the real-time positioning information of the reference drone includes: obtaining the successive operation path of the target drone and extracting the preset coordinate positions corresponding to each time point on the time axis of the successive operation path, the preset coordinate positions including planar position and altitude, the successive operation path referring to the remaining operation path segment planned but not yet executed by the target drone before the GPS positioning signal is lost; obtaining the real-time positioning information of the reference drone at each corresponding time point; and calculating the real-time virtual positioning of the target drone based on the historical relative positional relationship, the real-time virtual positioning including the virtual planar coordinates and virtual altitude of the target drone at each time point.

[0013] By employing the aforementioned technical solution, and using the real-time position of the reference UAV as a dynamic benchmark, the calculated results are ensured to be closely anchored in the real-world coordinate system. Secondly, by combining the target UAV's own preset path, the expected trends in heading and speed are provided for the calculation, ensuring that the calculated results conform to its mission planning. Most importantly, the "historical relative position relationship" is used as a correction parameter to calibrate the theoretical position based on the preset path in real time, compensating for deviations between the actual and theoretical formations caused by airflow, control errors, etc. This multi-information fusion calculation method generates a virtual position that is not an isolated point, but a series of coordinates corresponding to the time axis, forming a smooth and dynamically adjusted temporary trajectory. This allows the calculated trajectory to both follow the reference object and conform to its own mission logic, with accuracy and reliability far exceeding simple inertial calculations, ensuring stable flight attitude and accurate path for the target UAV during GPS loss.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of sending the temporary tracking trajectory to the target drone, the method further includes: at preset verification intervals, selecting the closest unobstructed drone from a preset set of radar verification drones to perform directional millimeter-wave radar scanning on the target drone; acquiring the scanning data returned by the radar verification drone and extracting the corresponding real-time spatial information of the target drone, the real-time spatial information including the azimuth angle, straight-line distance, and height difference of the target drone relative to the radar verification drone; retrieving the preset theoretical spatial information of the radar verification drone and the target drone at the current verification time from the theoretical data stored before the operation; comparing the real-time spatial information with the preset theoretical spatial information to determine the verification result; if the verification result is greater than a set error threshold, updating the temporary tracking trajectory according to the preset theoretical spatial information.

[0015] While the virtual trajectory calculated based on the reference UAV achieves high initial accuracy using the aforementioned technical solution, small errors accumulate over time, potentially leading to trajectory drift. To address this, the solution utilizes a "radar-calibrated UAV" for periodic active detection. Millimeter-wave radar provides high-precision relative position information (azimuth, distance, altitude difference) unaffected by GPS signals. By comparing the radar's "real-time spatial information" with pre-stored "pre-defined theoretical spatial information," the system can accurately detect the deviation between the virtual trajectory and the theoretical formation. Once the deviation exceeds a threshold, the system uses this precise deviation data to "update the temporary tracking trajectory," essentially performing a high-precision "pull-back" calibration on the drifting trajectory. This closed-loop "calculation-calibration-correction" working mode effectively suppresses the cumulative effect of errors, ensuring that even with prolonged GPS signal loss, the target UAV's flight trajectory remains within acceptable error ranges, significantly improving the robustness and tracking accuracy of the entire system.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of sending the temporary tracking trajectory to the target drone, the method further includes: when the GPS positioning signal of the target drone is restored, obtaining the real-time GPS positioning information and the current timestamp of the target drone; obtaining the target path node corresponding to the total operation path according to the current timestamp, and determining the path segment that has not been executed after the target path node as the follow-up path after restoration; and sending the follow-up path after restoration to the target drone.

[0017] By employing the aforementioned technical solution, if the target drone is directly commanded to return to the point before the signal loss when its GPS signal is restored, it may cause the drone to perform a dangerous, high-angle turnaround, wasting valuable operational time. This solution, however, accurately locates the drone's current progress on the overall mission timeline by acquiring the "real-time GPS positioning information" and "current timestamp" at the moment of recovery. Based on this progress, the system can intelligently extract "unexecuted path segments" from the "overall operational path" as the "follow-up path after recovery." This ensures that the drone does not need to backtrack but seamlessly integrates into the subsequent mission flow from its current position in the smoothest and most efficient way. This intelligent path replanning mechanism avoids sudden changes in flight attitude and flight path confusion caused by state transitions, ensuring operational continuity and overall efficiency, and achieving a smooth and safe transition from an abnormal state to a normal state.

[0018] In some embodiments, in conjunction with the first aspect, the method further includes: if the GPS positioning signal of the reference drone is also detected to be lost, then re-selecting the operational drones from the candidate operational drones with a spatial matching degree greater than a set spatial ratio to form a backup reference candidate pool; calculating the sum of the straight-line distances between each operational drone in the backup reference candidate pool and the target operational drone and the reference drone at the current time, and selecting the operational drone with the shortest sum of straight-line distances as the new reference drone; and generating temporary tracking trajectories for the target operational drone and the reference drone based on the new reference drone.

[0019] By employing the aforementioned technical solution, the system did not collapse even in the severe "double failure" scenario where the reference drone itself also lost its GPS signal. Instead, it immediately activated its contingency plan, searching for a new reference drone from the previously selected "backup reference candidate pool." A key design element of this selection criterion is the "shortest total straight-line distance," meaning the new reference drone is closest to the geometric center of the "small group" consisting of the two drones that lost their signals, providing reference for both simultaneously in the most efficient and stable manner. Subsequently, based on this new reference, the system generated temporary trajectories for each of the two drones that lost their signals. This design effectively controls the spread of the fault; even in the event of localized multi-point failures, it can quickly rebuild a stable local reference frame, maintaining the integrity of most of the formation. This ensures that the entire cluster can maintain basic formation and operational capabilities even in the face of continuous, multi-point signal interference, demonstrating extremely high system survivability and reliability.

[0020] In a second aspect, this application provides a server comprising: one or more processors and a memory; the memory being coupled to the one or more processors, the memory being used to store computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the server to perform the methods described in the first aspect and any possible implementation thereof.

[0021] Thirdly, this application provides a computer-readable storage medium including instructions that, when executed on a server, cause the server to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, this application provides a computer program product that, when run on a server, causes the server to perform the method described in the first aspect and any possible implementation thereof.

[0023] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0024] 1. By employing a technique that dynamically calculates the real-time virtual positioning of the target drone when its GPS signal is lost, by selecting a reference drone with a normal GPS signal and innovatively combining the stable "historical relative position relationship" between the two with the "real-time positioning information" of the reference drone, the technical problem of drones being unable to maintain formation effectively after losing their GPS signal is effectively solved. This can easily lead to deviation from the preset trajectory, collisions, or operational failures. Thus, even without its own GPS signal, the target drone can accurately maintain its relative position in the swarm through reference coordination, ensuring the continuity of the trajectory and the safety of the operation, and significantly improving the reliability and robustness of swarm flight.

[0025] 2. By employing a technique that discretizes the planned paths of candidate and target drones over a future period in a spatiotemporal manner, and quantitatively assesses their motion coordination by calculating the frequency at which their spatial distance at corresponding time points is less than a preset threshold, thus determining the "spatial matching degree," this technique effectively solves the technical problem in existing technologies where the selection of a reference drone is based solely on a single dimension such as the current physical distance, which fails to predict future motion trends and may result in the selection of a non-optimal companion drone that is about to part ways, leading to unstable reference relationships. This technique transforms the selection of reference drones from post-event evaluation to pre-event prediction, scientifically and quantitatively screening out highly kinematically coordinated reference objects, laying a solid foundation for subsequent high-precision trajectory calculation.

[0026] 3. By employing a closed-loop verification technique that periodically uses independent radar to calibrate the UAV, actively scans and acquires the target UAV's real-time high-precision relative spatial information, and compares it with the theoretical spatial information stored before the operation, thereby discovering and correcting the accumulated errors of the virtual trajectory, the technical problem of virtual trajectory calculated solely based on initial relationships drifting due to the continuous accumulation of small errors such as airflow and control delays in scenarios with long-term GPS signal loss is effectively solved. This achieves dynamic calibration of the calculated trajectory, effectively suppressing the cumulative effect of errors, and ensuring that the target UAV's flight trajectory can still be accurately constrained within the allowable range even when there is no GPS signal for a long time, greatly improving the long-term robustness and accuracy of the trajectory tracking system. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating a drone trajectory tracking method in an embodiment of this application;

[0028] Figure 2 This is another flowchart illustrating the drone trajectory tracking method in this application embodiment;

[0029] Figure 3 This is a schematic diagram of the physical device structure of a server in an embodiment of this application. Detailed Implementation

[0030] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0031] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0032] For ease of understanding, the method provided in this implementation is described in process below. Please refer to [link / reference]. Figure 1 This is a flowchart illustrating a drone trajectory tracking method in an embodiment of this application.

[0033] S101. Receive GPS positioning signals from all operating drones in real time;

[0034] During drone operations, the server needs to maintain a stable communication connection with each drone to continuously receive its GPS positioning signals. Each drone has a built-in GPS module that periodically (usually multiple times per second) receives satellite signals and calculates its real-time location. The drone then wirelessly transmits these GPS positioning signals to the server in a specific data format (including drone identification, timestamp, latitude and longitude, altitude, etc.).

[0035] The server associates these GPS positioning signals with the corresponding drone identifiers and stores them in the database, while simultaneously updating the drone's real-time location display in the monitoring system. For large-scale drone swarm operations, the server needs to have powerful data processing and concurrent processing capabilities to ensure that it can simultaneously receive and process GPS positioning signals from dozens or even hundreds of drones without data loss or delay.

[0036] S102. When the GPS positioning signal of any target drone is lost, determine the reference drone of the target drone and obtain the real-time positioning information of the reference drone.

[0037] The moment the server detects the loss of the target drone's GPS signal, it needs to determine a reliable reference drone. This process can be divided into four closely linked stages:

[0038] The first stage is basic information collection. Once the server determines that a drone is the target drone through its signal reception monitoring mechanism, it immediately initiates the information collection process. This monitoring mechanism includes confirming a loss if no valid GPS signal data is received for three consecutive signal cycles. First, the server retrieves the drone's overall operational path, which is typically stored in the server's task planning database and includes the three-dimensional coordinates for each time point. Second, the server records the precise timestamp of the signal loss, which serves as the benchmark for all subsequent time-related calculations. Finally, the server extracts the GPS location of the loss point. This location information is the spatial origin for searching candidate drones and is also an important reference point for distance calculation. For example, in a drone formation performance, if drone numbered D-08 loses its signal at 10:05:23.123, the server will immediately obtain its preset performance trajectory, the precise time, and its last position (e.g., longitude 116.3°, latitude 39.9°, altitude 100 meters).

[0039] The second stage is candidate drone screening. The server searches outwards layer by layer from the GPS location of the lost drone, according to a preset distance gradient (e.g., 50 meters → 100 meters → 200 meters). Within each layer, all operational drones with normal GPS signals (i.e., the server can continuously receive their positioning data) are screened. During this process, the server queries the current positioning information of each drone in real time and determines whether it is within the current search layer using a spatial distance calculation formula (a spherical distance formula based on latitude and longitude). If the number of candidate drones screened in the first layer (50 meters) has reached the set number (e.g., 8), the search range is stopped from expanding; if the number is insufficient, it automatically enters the next layer (100 meters) to continue searching until the number requirement is met. This mechanism ensures the spatial proximity of candidate drones while avoiding the problem of not finding enough candidates due to an excessively small search range. For example, in an agricultural plant protection scenario, if there are only 3 valid drones within 50 meters of the target drone's loss point, and the set number of candidates is 5, the server will automatically expand the search range to 100 meters until more than 5 candidate drones are found.

[0040] The third stage is spatial matching degree calculation. This is a crucial step in ensuring the path coordination between the reference UAV and the target UAV. Specifically, the server first extracts the path location point of the target UAV at the time stamp corresponding to the signal loss from the total operational path. Using this as a starting point, it extracts the expected path segment within a future set follow-up time (e.g., 30 seconds)—this path is the trajectory the target UAV would follow if it had not lost the signal. Simultaneously, based on the same timestamp, it obtains the planned path segments of each candidate UAV within the same follow-up time. Subsequently, the two paths are discretized into multiple coordinate points at preset time intervals (e.g., 1 second / point). For example, a 30-second path will generate coordinates for 30 time points. For each corresponding time point, the spatial distance (three-dimensional distance, combining horizontal and vertical differences) between the planned coordinates of the candidate UAV and the expected coordinates of the target UAV is calculated, and the number of time points with a distance less than a preset spatial threshold (e.g., 10 meters) is counted. Finally, the ratio of this number to the total number of time points is determined as the spatial matching degree. For example, if the distance between 21 of the 30 time points is less than 10 meters, the spatial matching degree is 70%. This calculation can effectively filter out candidates that are "on the same path" as the target drone, avoiding the selection of drones with significantly different paths as references.

[0041] The fourth stage involves determining the reference drone. The server first filters candidate drones with a spatial matching degree greater than a set proportion (e.g., 60%), which are considered to have high path coordination with the target drone. Then, the straight-line distance between these qualified candidates and the target drone's loss point at the moment of GPS signal loss is calculated (based on their positioning data at that moment, using the spherical distance formula). Finally, the drone with the shortest straight-line distance among these candidates is selected as the reference drone. This selection logic ensures both path consistency between the reference and target drones (avoiding position estimation errors due to path differences) and spatial proximity (the closer the distance, the more stable the relative positional relationship and the higher the estimation accuracy). For example, if after screening, three candidate drones have a spatial matching degree exceeding 60%, and their straight-line distances to the target drone are 35 meters, 28 meters, and 42 meters respectively, then the drone at a distance of 28 meters is selected as the reference.

[0042] The effect of this step is to find the optimal reference object for the target drone that has lost GPS signal. By filtering in multiple dimensions (spatial distance, path matching degree), the reliability and relevance of the reference drone are ensured, laying a solid foundation for the subsequent calculation of the target drone's position based on relative position relationship. This effectively reduces the risk of positioning deviation caused by improper reference selection and improves the stability of drone trajectory tracking in complex scenarios.

[0043] S103. Calculate the historical relative positional relationship between the reference UAV and the target UAV, including the horizontal relative distance, altitude difference and relative azimuth angle.

[0044] This step is performed immediately after the reference drone is identified. Its purpose is to establish a stable positional correlation model between the two drones using historical data, providing a calculation benchmark for subsequent position estimation.

[0045] The server needs to determine the time range of the historical data. Typically, 1-3 minutes before the target drone lost its GPS signal is selected (this can be dynamically adjusted according to the operational scenario; for example, in formation flying where the drone's relative position is stable, it can be shortened to 30 seconds; in agricultural plant protection where the relative position changes slowly, it can be extended to 5 minutes), and both the drone and the target drone must have valid GPS signals within this time period to ensure data integrity and reliability. For example, if the target drone loses its signal at 10:00:00, then the positioning data between 9:58:00 and 10:00:00 will be used for calculation.

[0046] Secondly, extract all GPS positioning records of the two drones within this time period from the database. Each record includes a timestamp (accurate to milliseconds), longitude, latitude, and altitude. To ensure data consistency, the timestamps need to be aligned—if the timestamps are not completely consistent, linear interpolation is used to generate positioning data at the same time point (e.g., if the reference drone has data at t1 and t3, and the target drone has data at t2, the positions of both can be calculated at time t2 through interpolation).

[0047] Next, we will calculate the three core parameters respectively:

[0048] Horizontal relative distance: First, convert the latitude and longitude coordinates to Cartesian coordinates (x, y) using UTM projection to eliminate the influence of the Earth's curvature; then calculate using the planar distance formula: For example, if the reference drone coordinates after conversion are (1000, 2000) and the target drone is (1050, 2050), then the horizontal distance is approximately 70.7 meters.

[0049] Altitude difference: Obtained directly by subtracting the altitudes: Altitude difference = Reference drone altitude - Target drone altitude. If the reference drone's altitude is 120 meters and the target drone's altitude is 110 meters, then the altitude difference is 10 meters.

[0050] Relative azimuth: Establish a coordinate system with the reference UAV as the origin (positive y-axis is due north, positive x-axis is due east), calculate the relative coordinates of the target UAV (Δx = xtarget - xreference, Δy = ytarget - yreference); then calculate the angle using the arctangent function: azimuth = arctan2(Δx, Δy) × (180 / π). If the result is negative, add 360° to convert it to a range of 0°-360°. For example, when Δx = 50° and Δy = 50°, the azimuth is 45° (northeast).

[0051] Finally, statistical processing was performed on the calculated sets of parameters to obtain stable historical relative positional relationships.

[0052] The effect of this step is to establish a "spatial correlation benchmark" between the reference drone and the target drone through historical data mining, avoiding the errors that may be caused by directly using the relative position at a single moment.

[0053] S104. Based on the historical relative position relationship and the real-time positioning information of the reference UAV, calculate the real-time virtual positioning of the target operation UAV, and generate a temporary tracking trajectory based on the real-time virtual positioning.

[0054] This step is performed after the target drone loses its GPS signal and a reference drone has been identified. It is a crucial process for calculating the virtual location of the target drone. The purpose is to determine its approximate location by referring to the location and historical relative relationship of the reference drone when the target drone cannot obtain its own GPS signal.

[0055] First, the server obtains the successive operation path of the target drone and extracts the preset coordinates for each time point. After the target drone loses its GPS signal, the server immediately retrieves the drone's total operation path from the operation planning system. Then, starting from the time stamp of signal loss, it extracts the path segment after that moment; this segment is the successive operation path. For example, if the target drone's total operation path is from point A to point B and then to point C, and the signal is lost before reaching point B, then the path from the location at the time of signal loss to point C is the successive operation path. Next, the server discretizes the successive operation path on the time axis according to preset time intervals (this interval can be set according to the operation accuracy requirements; for high-precision scenarios such as formation performances, it can be set to 0.1 seconds, and for general scenarios such as agricultural plant protection, it can be set to 1 second), obtaining a series of time points, and extracting the preset coordinates for each time point, including latitude and longitude (planar position) and altitude. These preset coordinates are the positions the target drone should reach under normal circumstances (when the GPS signal is not lost), and are an important reference for subsequent verification of the rationality of virtual positioning.

[0056] Secondly, the server acquires the real-time positioning information of the reference drone at each corresponding time point. After extracting the time points of the target drone's succession path, the server receives GPS positioning signals sent by the reference drone in real time and filters out the positioning information corresponding to these time points. If the time interval between the positioning signals sent by the reference drone is not exactly the same as the time interval of the target drone's succession path, the server uses interpolation to process the reference drone's positioning data to obtain accurate real-time positioning information for each corresponding time point. For example, if the time points of the target drone's succession path are t1, t2, t3, ..., and the reference drone has positioning data at times t1-Δt and t1+Δt, the server will calculate the reference drone's positioning information at time t1 through linear interpolation. This real-time positioning information includes the latitude, longitude, and altitude of the reference drone at each corresponding time point and serves as the benchmark data for estimating the virtual positioning of the target drone.

[0057] Finally, the real-time virtual positioning of the target drone is calculated based on historical relative position relationships. This process involves calculations of both planar coordinates and altitude. For the calculation of virtual planar coordinates, the server extrapolates based on the horizontal relative distance and relative azimuth angle from historical relative position relationships, using the planar coordinates of the reference drone at the corresponding time point as a benchmark. Specifically, using the planar coordinates of the reference drone as the origin, the direction of the target drone relative to the reference drone is determined based on the relative azimuth angle. Then, combined with the horizontal relative distance, the virtual planar coordinates (latitude and longitude) of the target drone are calculated using trigonometric functions. For example, if the latitude and longitude of the reference drone at a certain time point is (x0, y0), the historical relative azimuth angle is 30°, and the horizontal relative distance is 50 meters, then the virtual planar coordinates (x, y) of the target drone can be calculated using x = x0 + 50 × sin(30°) and y = y0 + 50 × cos(30°) (in actual calculations, the conversion relationship between latitude / longitude and distance needs to be considered). To calculate the virtual altitude, the server uses the altitude difference in historical relative positions. It subtracts this altitude difference from the real-time altitude of the reference drone at the corresponding point in time to obtain the virtual altitude of the target drone. For example, if the real-time altitude of the reference drone at a certain point in time is 100 meters, and the historical altitude difference is 5 meters (i.e., the reference drone is 5 meters higher than the target drone), then the virtual altitude of the target drone is 95 meters.

[0058] During the calculation process, the server also compares the calculated real-time virtual location with the preset coordinates of the corresponding time point in the subsequent operation path. If the deviation between the two is within a preset reasonable range (such as 10 meters), the virtual location is considered valid; if the deviation is too large, the server will re-check whether the calculation of the historical relative position relationship is accurate, or verify the real-time positioning information of the reference drone, and adjust the calculation parameters if necessary to improve the accuracy of the virtual positioning.

[0059] The effect of this step is that, in the event that the target drone loses its GPS signal, it obtains its real-time virtual positioning through scientific calculation methods, providing accurate location data for generating a temporary tracking trajectory. This ensures that the target drone can continue to fly along a path close to the original plan, effectively avoiding problems such as trajectory deviation and operation interruption caused by positioning loss, and ensuring the safety and continuity of drone operations.

[0060] S105. Send the temporary tracking trajectory to the target drone.

[0061] First, the server needs to format the generated temporary tracking trajectory. Since the drone's flight control system can only recognize data in specific formats, the server converts the temporary tracking trajectory into a format understandable by the target drone. The server then sends out the formatted temporary tracking trajectory via a wireless communication link (such as 4G / 5G, a dedicated radio, etc.) with the target drone. Upon receiving the temporary tracking trajectory and confirming the data is correct, the target drone uses it as its current flight command. The flight control system calculates the deviation between its current position and the target position in real time based on the position information in the temporary tracking trajectory, and corrects its flight attitude by adjusting motor speeds, servo angles, etc., to ensure the drone flies along the temporary tracking trajectory.

[0062] During the drone's flight, the server continuously communicates with the target drone, receiving its flight status information (such as current flight speed and attitude angle), and uses this information to determine whether the drone is strictly following the temporary tracking trajectory. If the drone deviates from the temporary tracking trajectory by more than a set threshold (such as 3 meters), the server will recalculate and generate a new temporary tracking trajectory, and send it back to the target drone to ensure that it always flies within a reasonable trajectory range.

[0063] In the above embodiment, by receiving GPS positioning signals from all operating drones in real time to fully grasp their position status, a reference drone can be quickly identified and its positioning information obtained when the target drone loses its GPS signal. The real-time virtual positioning of the target drone is calculated by combining historical relative position relationships and a temporary tracking trajectory is generated and sent out, forming a complete positioning substitution and trajectory control closed loop. Therefore, it can continuously provide reliable positioning reference and flight guidance even when the target drone loses its GPS signal, effectively solving problems such as drone trajectory deviation, cluster collision or operation interruption caused by GPS signal loss, and thus achieving the safety, continuity and accuracy of drone operations in complex scenarios.

[0064] In some embodiments, after step S105, the server continuously monitors the GPS signal status of the target drone. When stable and valid GPS positioning data is received for multiple consecutive signal cycles (e.g., 3 cycles, each 500 milliseconds), it is determined that the GPS positioning signal has been restored. At this time, the server immediately acquires two key pieces of information: first, the real-time GPS positioning information of the target drone, which is the drone's current actual location and is used for subsequent comparison with theoretical path nodes; second, the current timestamp, which serves as the core benchmark for matching the overall operation path, ensuring that accurate target path nodes are found.

[0065] Secondly, the follow-up path after recovery is determined. The server retrieves the theoretical path node corresponding to the current timestamp from the preset overall task path database—the "target path node." For example, if the overall task path records "timestamp 10:00:00, path node is point A (longitude 116.3°, latitude 39.9°, altitude 100 meters)," and the current timestamp of signal recovery happens to be this value, then point A is the target path node. If the current timestamp is not at a preset time point on the overall task path (e.g., the overall task path sets nodes at 1-second intervals, and the recovery timestamp is between two nodes), the server will use linear interpolation to calculate the theoretical location corresponding to that timestamp, which will then be used as the target path node. Next, the server extracts all unexecuted path segments after the target path node from the overall operation path and identifies them as the follow-up path after recovery. This path is a continuation of the original operation plan, ensuring that the UAV can complete the remaining tasks. For example, if the original overall operation path is A→B→C→D, and the target path node is B when the signal is restored, then the follow-up path after recovery will be B→C→D. Finally, the follow-up path after recovery is sent out. The server first sends the path data to the target UAV in a format recognizable by the UAV's flight control system. After receiving the data, the target UAV compares its current real-time GPS positioning information with the deviation from the target path node. If the deviation is small (e.g., less than 3 meters), it directly switches to flying according to the follow-up path after recovery; if the deviation is large (e.g., greater than 3 meters), it automatically generates a short path that smoothly transitions from the current actual position to the target path node, and then connects to the follow-up path after recovery to avoid flight instability caused by sudden attitude changes.

[0066] In the above embodiments, a "seamless connection" of UAV operations after GPS signal recovery is achieved, which effectively solves the problem that UAVs may deviate from the original operation path after signal recovery, resulting in operation omissions or repeated operations. It ensures that UAVs can accurately return to the original planned tasks, guarantees the continuity and integrity of the entire operation process, and improves the reliability of UAV operations.

[0067] In some embodiments, after step S105, in order to avoid the accumulation of virtual positioning errors caused by the target operation drone relying on the relative position calculation of the reference drone for a long time, and to ensure that the temporary tracking trajectory always conforms to the accuracy requirements of the original operation plan, a dynamic verification and trajectory correction mechanism can be introduced, such as by executing steps S201-S205.

[0068] Based on the above, the following is a more detailed description of the process provided in this implementation. Please refer to [link / reference]. Figure 2 This is another flowchart illustrating the drone trajectory tracking method in this application.

[0069] S201. Every preset verification time, select the nearest unobstructed operational drone from the preset radar verification drones to perform directional millimeter-wave radar scanning on the target operational drone.

[0070] This step is performed after the target drone loses its GPS signal and the server has generated a temporary tracking trajectory. It is a periodic verification mechanism for virtual positioning. The purpose is to verify the accuracy of virtual positioning through independent radar detection and avoid the accumulation of errors caused by long-term reliance on relative position calculations.

[0071] First, the server triggers the verification process according to a preset verification interval (e.g., every 30 seconds). After triggering, the server filters qualified candidates from a preset list of radar verification drones: First, based on the real-time virtual positioning of the target drone, the server calculates the straight-line distance (based on three-dimensional distance using latitude, longitude, and altitude) between each radar verification drone and the virtual positioning location; Second, the server selects the closest drones (e.g., 3 drones) as preliminary candidates; Third, the server uses an obstacle detection algorithm (combined with three-dimensional map data of the work area, such as building height and terrain undulations) to determine whether there is any obstruction between the preliminary candidate drones and the target drone—if there is an object higher than the height of the line connecting the two drones on the radar signal propagation path, it is considered an obstruction and is excluded; Finally, the server selects the closest drone from the unobstructed candidates as the drone to be used for this verification.

[0072] Once selected, the server sends a scanning command to the radar calibration drone, including the real-time virtual positioning of the target drone (as a reference for the scanning direction) and scanning parameters (such as radar power and scan duration). Upon receiving the command, the radar calibration drone adjusts its attitude to align its radar antenna with the virtual positioning direction of the target and initiates a directional millimeter-wave radar scan. Millimeter-wave radar has the ability to penetrate smoke and light rain and is highly resistant to electromagnetic interference, making it suitable for short-range positioning in complex scenarios. The scanning process lasts 1-2 seconds to ensure sufficient reflected signal data is acquired.

[0073] The effect of this step is to provide an objective external verification method for virtual positioning by periodically activating an independent radar verification mechanism, thus avoiding the cumulative errors that may occur from relying solely on relative position calculations.

[0074] S202. Obtain the scanning data returned by the radar verification UAV and extract the real-time spatial information of the target operation UAV. The real-time spatial information includes the azimuth angle, straight-line distance and altitude difference of the target operation UAV relative to the radar verification UAV.

[0075] After completing the directional scan, the radar-calibrated UAV sends the scan results, including the raw radar echo data, to the server via an encrypted communication link. The raw data includes: signal transmission time, reflected signal reception time, signal frequency offset, signal strength, etc.—this data contains information about the target UAV's distance and motion status.

[0076] After receiving the scan data, the server initiates the data parsing process:

[0077] Calculating the straight-line distance: Based on the propagation speed of radar signals (speed of light) and the round-trip time difference (reception time - transmission time), the straight-line distance between the two UAVs is calculated using the formula "Distance = (Speed ​​of light × Time difference) / 2" (eliminating the effect of the doubled distance traveled by the radar signal). For example, the round-trip time difference is 1.33 × 10⁻⁻⁴. 7 If the time is approximately 40 meters, then the straight-line distance is approximately 20 meters.

[0078] Calculate the azimuth angle: Determine the horizontal azimuth angle of the target UAV relative to the radar-calibrated UAV by using the beam pointing angle of the radar antenna (which is aligned with the virtual positioning direction of the target during scanning) and the angular offset of the reflected signal (calculated from the phase difference of the radar array antenna). For example, if the antenna initially points at 30° and the signal offset is 5°, then the actual azimuth angle is 35°.

[0079] Calculating the height difference: Combining the radar's elevation angle (the angle between the antenna and the horizontal plane) and the straight-line distance, the vertical distance is calculated using trigonometric functions: Height difference = Straight-line distance × sin(elevation angle). If the straight-line distance is 20 meters and the elevation angle is 15°, then the height difference ≈ 5.18 meters (a positive value indicates the target UAV is higher, and a negative value indicates it is lower). During the analysis process, the server will filter the data.

[0080] S203. Retrieve the preset theoretical spatial information of the radar verification UAV and the target operation UAV at the current verification time from the theoretical data stored before the operation.

[0081] This step provides a "theoretical baseline" based on the original job plan, which is used to compare with the real-time spatial information measured in reality to determine whether the virtual positioning deviates from the planned intent.

[0082] Before the task begins, the server generates detailed theoretical data based on the total operational path of all drones. For each time point (accurate to the second), it calculates the theoretical relative position between any two drones—including azimuth, straight-line distance, and altitude difference—and stores it in the database using a "time-drone pair" index structure. For example, in a formation performance task, the server pre-stores the theoretical distance (5 meters) and azimuth (90°) between drones D-01 and D-02 at each moment.

[0083] When the current verification time is reached, the server retrieves the corresponding data based on the following information:

[0084] The unique identifier of the UAV used in this radar verification (e.g., number R-03);

[0085] The unique identifier of the target drone (e.g., number D-08);

[0086] The exact timestamp of the current verification (e.g., 10:05:30.000).

[0087] The server retrieves records matching the above three elements from the theoretical database and extracts the corresponding preset theoretical spatial information, including the theoretical azimuth angle (e.g., 45°), theoretical straight-line distance (e.g., 25 meters), and theoretical altitude difference (e.g., 0 meters, flying at the same altitude) between R-03 and D-08 at that moment.

[0088] In this way, even if the relative relationship between the virtual positioning and the reference drone is stable, if both deviate from the original planned path, the problem can be found by comparing with theoretical data, providing a basis for subsequent trajectory correction.

[0089] S204. Compare the real-time spatial information with the preset theoretical spatial information to determine the verification result;

[0090] This step, performed after acquiring real-time spatial information and preset theoretical spatial information, is the core judgment step in the verification process. Its purpose is to quantitatively assess the degree of deviation between the actual position of the target drone and the theoretical plan, and to provide a basis for decision-making on whether the trajectory needs to be corrected.

[0091] The server compares real-time spatial information with preset theoretical spatial information item by item:

[0092] Azimuth deviation: Calculate the difference between the real-time azimuth and the theoretical azimuth (take the absolute value; if it exceeds 180°, subtract the value from 360° to ensure the deviation is within the range of 0-180°). For example, if the real-time azimuth is 48° and the theoretical azimuth is 45°, the deviation is 3°.

[0093] Straight-line distance deviation: Calculates the absolute difference between the real-time distance and the theoretical distance. For example, if the real-time distance is 28 meters and the theoretical distance is 25 meters, the deviation is 3 meters.

[0094] Height difference deviation: Calculates the absolute difference between the real-time height difference and the theoretical height difference. For example, if the real-time height difference is -1 meter and the theoretical height difference is 0 meters, the deviation is 1 meter.

[0095] Then, the three deviations are weighted and summed according to preset weights (set according to the accuracy requirements of each dimension in the operation scenario) to obtain the comprehensive verification result. For example, in formation performance, the azimuth accuracy requirement is the highest (weight 0.5), followed by distance (0.3), and the height difference is the lowest (0.2). Then the comprehensive error = 3°×0.5 + 3m×0.3 + 1m×0.2 = 1.5 + 0.9 + 0.2 = 2.6 (the unit is the comprehensive error value, which must be consistent with the set threshold unit).

[0096] In addition, the server will also perform threshold checks on individual deviations (such as azimuth deviation exceeding 10° or distance deviation exceeding 5 meters, which are judged as significant deviations). Even if the overall error does not exceed the standard, an alarm will be triggered if the individual deviation is too large, to ensure that all dimensions meet the accuracy requirements.

[0097] S205. If the verification result is greater than the set error threshold, the temporary tracking trajectory is updated according to the preset theoretical spatial information.

[0098] Among them, "setting error threshold" refers to the maximum allowable deviation range preset by the server. If the deviation exceeds this range, the virtual positioning is considered unreliable and needs to be corrected. "Preset theoretical spatial information" refers to the theoretical relative positional relationship between the target UAV and the radar verification UAV at the current moment, which is stored before the operation.

[0099] This step is mainly to avoid job failure caused by long-term accumulated errors when the virtual positioning deviates too much from the theoretical plan, by forcing a regression to the theoretical path.

[0100] The server determines whether the verification result exceeds a set error threshold (e.g., a comprehensive error value of 3.0). If it does, the trajectory update process is initiated.

[0101] Theoretical virtual positioning calculation: Based on the real-time positioning of the radar-calibrated UAV, and combined with preset theoretical spatial information (theoretical azimuth, distance, and altitude difference), the theoretical virtual positioning of the target UAV is calculated in reverse—that is, "the position where the target UAV should be at this moment if strictly following the operation plan." For example, if the real-time positioning of the radar-calibrated UAV is (x0, y0, h0), the theoretical azimuth is 45°, the distance is 25 meters, and the altitude difference is 0 meters, then the theoretical positioning of the target UAV is (x0 + 25 × sin45°, y0 + 25 × cos45°, h0).

[0102] Generate transition trajectory segment: To avoid drastic changes in the attitude of the drone caused by sudden changes in trajectory, the server generates a smooth transition trajectory segment between the current virtual positioning and the theoretical virtual positioning (such as using a cubic Bézier curve). The transition time is set according to the magnitude of the deviation (e.g., a 5-meter deviation corresponds to a 2-second transition, ensuring that the acceleration is within a safe range).

[0103] Update subsequent trajectory: Connect the transition trajectory segment with the remaining theoretical path segment in the original follow-up operation path to form a new temporary tracking trajectory. After the transition is completed, the new trajectory will strictly follow the theoretical path of the operation plan and will no longer rely on the relative position calculation of the reference UAV.

[0104] Simultaneously, the server records the reason for this trajectory update (e.g., "distance deviation 4 meters, overall error 3.2 > 3.0"), the time, and the adjustment range, serving as a basis for subsequent operation analysis. If the verification result does not exceed the threshold, the original temporary tracking trajectory remains unchanged, and positioning continues to be calculated by referencing the drone. For example, in a drone fireworks display, if the verification finds that the target drone deviates from its theoretical position by 6 meters (overall error 3.5 > 3.0), the server will immediately calculate its theoretical position and generate a 2-second transition trajectory to allow the drone to smoothly return to the theoretical path, ensuring the integrity of the fireworks pattern.

[0105] In this embodiment, by periodically activating the radar verification mechanism, the nearest and unobstructed radar verification drone is selected to perform millimeter-wave radar scanning on the target drone that has lost GPS signal to obtain real-time spatial information. This information is then accurately compared with the preset theoretical spatial information stored before the operation. When the verification result exceeds the set error threshold, the temporary tracking trajectory is updated in a timely manner, forming a closed loop of dynamic verification and trajectory correction for virtual positioning. Therefore, it can effectively avoid the accumulation of errors caused by long-term reliance on relative position calculation, effectively solve the problem of the drone's virtual positioning deviation gradually increasing and deviating from the original operation plan after the GPS signal is lost, and thus achieve continuous and accurate control of the drone's operation trajectory in complex scenarios, ensuring the accuracy and safety of the operation.

[0106] In some embodiments, if it is determined that the GPS signal of the target drone has been lost and the original reference drone also subsequently loses its GPS signal, in order to avoid the crisis of dual positioning loss caused by the failure of the reference drone, a new reference drone can be quickly selected to achieve synchronous trajectory control of the target drone and the original reference drone, thereby ensuring operational safety.

[0107] First, the triggering conditions and backup candidate pool are established. The server monitors the GPS signal status of the original reference drone in real time. When a signal loss is detected (the judgment criteria are the same as for the target drone, such as no valid data for 3 consecutive cycles), the emergency screening process is immediately initiated. Since candidate drones with "spatial matching degree greater than the set spatial ratio" have already been screened when the target drone first loses its signal (such as the 5 path-matching drones screened in S301), the server directly removes individuals with invalid GPS signals from this group of drones. The remaining drones with normal signals constitute the backup reference candidate pool. This avoids the time delay of re-searching the entire range and ensures the speed of emergency response. For example, if 2 of the original 5 candidate drones have normal signals, the backup candidate pool includes these 2 drones.

[0108] Secondly, the selection of a new reference drone. The server first obtains two key location information: one is the current real-time virtual location of the target drone (calculated based on the relative position of the original reference drone before it was lost); the other is the last valid location of the original reference drone when its signal was lost. Then, it calculates the "sum of straight-line distances" between each drone in the backup candidate pool and these two locations: total distance = (straight-line distance between the candidate drone and the target drone's virtual location) + (straight-line distance between the candidate drone and the original reference drone's last location). The drone with the shortest total distance is selected as the new reference drone—the core of this selection logic is to ensure that the new reference drone is simultaneously close to both the target drone and the original reference drone, so that the target drone's position can be stably calculated, and the lost signal of the original reference drone can be simultaneously controlled, preventing both from going out of control. For example, if the total distance of candidate drone A is 40 meters and that of candidate drone B is 55 meters, then A is selected as the new reference drone.

[0109] Finally, temporary tracking trajectories for both drones are generated simultaneously. Based on the real-time positioning information of the new reference drone, the server calculates the trajectory using two sets of historical relative positional relationships: first, the historical relative positional relationship between the target drone and the new reference drone (if there is no direct historical data, it can be calculated from the relative positions of the target drone and the original reference drone, and the original reference drone and the new reference drone), generating a temporary tracking trajectory for the target drone; second, the historical relative positional relationship between the original reference drone and the new reference drone (similarly, if there is no direct data, it is obtained through calculation), generating a temporary tracking trajectory for the original reference drone. Both trajectories are formatted according to the previous requirements (including timestamps, coordinates, and control parameters) and then sent to the target drone and the original reference drone respectively. If the original reference drone cannot receive commands due to signal loss, the server uses the millimeter-wave radar scan of the new reference drone (such as the S201-S205 logic) to calibrate its position in real time, ensuring effective trajectory control.

[0110] In the above embodiments, a dual emergency protection mechanism for "reference drone failure" is constructed, which effectively solves the problem of both the target drone and the original reference drone losing control after the original reference drone loses its signal. By quickly selecting a new reference drone and simultaneously generating dual-drone trajectories, continuous and stable control of the drone swarm in complex scenarios is achieved, further improving the reliability of the system in dealing with extreme signal loss scenarios.

[0111] The server in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference]. Figure 3 This is a schematic diagram of the physical device structure of a server in an embodiment of this application.

[0112] It should be noted that, Figure 3 The server structure shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0113] like Figure 3 As shown, the server includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on a program stored in Read-Only Memory (ROM) 302 or a program loaded from storage portion 308 into Random Access Memory (RAM) 303, such as performing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0114] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0115] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the various functions defined in the present invention.

[0116] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0117] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0118] Specifically, the server in this embodiment includes a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, it implements the UAV trajectory tracking method provided in the above embodiment.

[0119] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the server described in the above embodiments; or it may exist independently and not assembled into the server. The storage medium carries one or more computer programs that, when executed by a processor of the server, cause the server to implement the UAV trajectory tracking method provided in the above embodiments.

[0120] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0121] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0122] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for tracking the trajectory of a drone, applied to a server, characterized in that, The method includes: It receives GPS positioning signals from all operational drones in real time. When the GPS positioning signal of any target drone is lost, the reference drone of the target drone is determined and the real-time positioning information of the reference drone is obtained. Calculate the historical relative positional relationship between the reference UAV and the target UAV, wherein the historical relative positional relationship includes horizontal relative distance, altitude difference, and relative azimuth angle; Based on the historical relative position relationship and the real-time positioning information of the reference UAV, the real-time virtual positioning of the target UAV is calculated, and a temporary tracking trajectory is generated based on the real-time virtual positioning. The temporary tracking trajectory is sent to the target drone. The step of determining a reference drone when the GPS positioning signal of any target drone is detected to be lost includes: When the GPS positioning signal of any target drone is detected to be lost, the total operation path, the timestamp of signal loss, and the GPS location of the last valid point of loss when the signal was lost are obtained. Based on the GPS positioning of the lost point, a search is performed outwards according to a preset distance gradient to filter out candidate drones with valid GPS positioning signals that reach a set number of candidates. Based on the total operation path and the timestamp, calculate the spatial matching degree between each candidate operation drone and the target operation drone; Candidate drones with a spatial matching degree greater than a set spatial ratio are selected, and the straight-line distance between each candidate drone and the target drone at the moment of GPS signal loss is calculated. The drone with the shortest straight-line distance is selected as the reference drone.

2. The method according to claim 1, characterized in that, The step of calculating the spatial matching degree between each candidate operational drone and the target operational drone based on the total operation path and the timestamp includes: Extract the path location points of the target drone at the timestamp from the total operation path; Starting from the path location point, extract the expected path segment that the target drone should execute within the set continuation time. Based on the timestamp, the planned path segment of each candidate UAV during the set consecutive time period is obtained; The expected path segment and each of the planned path segments are discretized into coordinate points of multiple time points according to a preset time interval; Calculate the spatial distance between the planned path segment and each expected path segment at each corresponding time point, and count the number of time points where the spatial distance is less than a preset spatial threshold; The ratio of the number of time points to the total number of time points within the set consecutive time period is determined as the spatial matching degree.

3. The method according to claim 1, characterized in that, The step of calculating the real-time virtual location of the target drone based on the historical relative position relationship and the real-time positioning information of the reference drone includes: Obtain the successive operation path of the target operation drone, and extract the preset coordinate positions corresponding to each time point on the time axis of the successive operation path. The preset coordinate positions include planar position and altitude. The successive operation path refers to the remaining operation path segment that the target operation drone has planned but has not yet executed before the GPS positioning signal is lost. Obtain the real-time positioning information of the reference UAV at each corresponding time point; Based on the historical relative positional relationships, the real-time virtual positioning of the target operation drone is calculated. The real-time virtual positioning includes the virtual planar coordinates and virtual altitude of the target operation drone at each time point.

4. The method according to claim 1, characterized in that, After the step of sending the temporary tracking trajectory to the target drone, the method further includes: At preset verification intervals, select the nearest unobstructed operational drone from the preset radar verification drones to perform directional millimeter-wave radar scanning on the target operational drone. The scanning data returned by the radar verification UAV is obtained, and the real-time spatial information of the target operation UAV is extracted. The real-time spatial information includes the azimuth angle, straight-line distance and altitude difference of the target operation UAV relative to the radar verification UAV. Retrieve the preset theoretical spatial information of the radar verification UAV and the target operation UAV at the current verification time from the theoretical data stored before the operation; The real-time spatial information is compared with the preset theoretical spatial information to determine the verification result; If the verification result is greater than the set error threshold, the temporary tracking trajectory is updated according to the preset theoretical spatial information.

5. The method according to claim 1, characterized in that, After the step of sending the temporary tracking trajectory to the target drone, the method further includes: When the GPS positioning signal of the target drone is restored, the real-time GPS positioning information and the current timestamp of the target drone are obtained. The target path node in the total job path is obtained according to the current timestamp, and the path segment that has not been executed after the target path node is determined as the follow-up path after recovery. The restored follow-up path is sent to the target operating drone.

6. The method according to claim 1, characterized in that, Also includes: If the GPS positioning signal of the reference drone is also lost, the operational drones are re-selected from the candidate operational drones with a spatial matching degree greater than the set spatial ratio to form a backup reference candidate pool. Calculate the total straight-line distance between each operational drone in the backup reference candidate pool and the target operational drone and the reference drone at the current time, and select the operational drone with the shortest total straight-line distance as the new reference drone; Based on the new reference UAV, temporary tracking trajectories are generated for the target UAV and the reference UAV, respectively.

7. A server, characterized in that, The server includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, and the one or more processors invoking the computer instructions to cause the server to perform the method as described in any one of claims 1-6.

8. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the server, it causes the server to perform the method as described in any one of claims 1-6.

9. A computer program product, characterized in that, When the computer program product is run on the server, the server performs the method as described in any one of claims 1-6.

Images