Historical flight path information-based swarm unmanned aerial vehicle hovering prediction method

Abstract

The invention discloses a swarm unmanned aerial vehicle hovering prediction method based on historical track information, and relates to the technical field of radar target tracking. The method comprises the following steps: S1, determining a speed threshold value and a radar sampling period threshold value; s2, comparing the speed of each unmanned aerial vehicle with a speed threshold value, and marking the unmanned aerial vehicle which may hover or removing the mark of the unmanned aerial vehicle which may hover; s3, acquiring whether the track corresponding to each unmanned aerial vehicle at the next moment obtains an associated trace point and the mark of the unmanned aerial vehicle which may hover, and judging whether to terminate the track or not; and S4, carrying out flight path termination on the flight path corresponding to the unmanned aerial vehicle which exceeds the radar scanning period threshold and is not subjected to trace point updating. According to the swarm unmanned aerial vehicle hovering prediction method based on the historical flight path information provided by the invention, the problems of flight path breakage and redundant flight path increase caused by a flight path termination algorithm in a traditional radar target tracking process are solved, and the tracking effect of a radar on the swarm unmanned aerial vehicle is improved.

Description

Technical Field

[0001] The invention relates to the technical field of radar target tracking, and in particular to a hovering prediction method for a swarm of unmanned aerial vehicles based on historical track information. Background Art

[0002] In the field of modern aviation technology and radar detection, swarm drones play an increasingly important role in military reconnaissance, disaster monitoring, logistics and distribution, etc., due to their collaborative operation and flexible deployment. However, the complex flight state of swarm drones in the air, especially the hovering state, poses a severe challenge to radar tracking technology.

[0003] In the radar detection principle, usually only targets that reach a certain speed will be clearly displayed on the radar. UAVs have the ability to hover in the air. When some of the drones in the swarm are in a hovering state, the traditional track termination algorithm exposes serious defects. Once the traditional algorithm is used, the track corresponding to the hovering drone will be terminated incorrectly. When these hovering drones resume movement, it is necessary to restart the track start phase and generate a new track. This series of operations causes the track results finally obtained to be broken, which destroys the continuity of tracking and makes it difficult to accurately grasp the overall picture of the drone's flight trajectory; at the same time, redundant tracks also increase in large quantities, interfering with the radar's processing of effective information, greatly affecting the radar's tracking accuracy and effect of the swarm drone, and unable to meet the strict requirements of practical applications such as real-time and accurate tracking of targets in military operations and continuous and stable grasp of drone positions during disaster monitoring.

[0004] At present, with the continuous expansion of the application scope of swarm drones and the continuous increase in the frequency of use, the demand for efficient and accurate tracking technology is becoming more and more urgent. At present, most radars are generally equipped with moving target display devices, but commonly used rotor drones can hover in the air, which will cause the radar to lose the point track information of the drone. For the traditional data interconnection algorithm, the track corresponding to the drone cannot obtain the associated point track at the next moment, resulting in entering the track termination stage, which seriously affects the quality of the final track. An innovative algorithm is urgently needed to improve the radar's tracking performance for swarm drones to ensure stable and accurate tracking in various complex situations. Summary of the invention

[0005] The purpose of the present invention is to propose a hovering prediction method for swarm UAVs based on historical track information, which effectively solves the problems of track breakage and increased redundant tracks that occur in traditional track termination algorithms when facing hovering UAVs, and significantly improves the tracking effect of radar on swarm UAVs.

[0006] To achieve the above object, the present invention proposes a hovering prediction method for a swarm of UAVs based on historical track information, and the specific steps are as follows: Step S1: Setting speed threshold and radar sampling period threshold ; Step S2: Set the speed and speed threshold of each drone Compare and mark possible hovering drones or remove marks of possible hovering drones; Step S3: Obtain the track corresponding to each UAV at the next moment to see whether the associated point track is obtained and mark the UAV that may be hovering, and determine whether to terminate the track; Step S4: Exceeding the radar sampling period threshold The track corresponding to the drone without point track update is terminated.

[0007] Preferably, in step S2, after the state estimation phase at each moment is completed, the speed of each drone at that moment is recorded. , and compare it with the speed threshold For comparison, if and the drone has not been marked as a possible hovering drone, then mark the drone as a possible hovering drone; if And if the drone has been marked as a possible hovering drone, remove the mark of the drone.

[0008] Preferably, in step S3, if the track corresponding to a certain UAV at the next moment is not associated with the measurement data, check whether this UAV is marked as a possible hovering UAV. If it is marked as a possible hovering UAV, continue to retain the track for data interconnection and state estimation; if this UAV is not marked as a possible hovering UAV, directly transfer the track to the track termination stage.

[0009] Preferably, in step S4, if a drone corresponding to a track is marked as a drone in a possible hovering state and the track has continuous If there is no update during the radar sampling period, the track will be transferred to the track termination stage.

[0010] Therefore, the present invention proposes a hovering prediction method for a swarm UAV based on historical track information, and its beneficial effects are as follows: (1) The present invention proposes a method for predicting hovering of swarm UAVs based on historical track information. By setting a speed threshold and a radar sampling period threshold, the method combines the comparison result between the UAV speed and the threshold to mark or remove UAVs that may be hovering. In the track association stage, the track of UAVs that may be hovering and not associated with measurement data will not be terminated blindly, but will be retained for subsequent processing. This mechanism avoids the erroneous termination of the track due to the hovering of the UAV, ensures the continuity and accuracy of the track, fundamentally reduces the occurrence of track breakage, and enables the radar to track the UAV target more accurately.

[0011] (2) The hovering prediction method of swarm UAVs based on historical track information proposed in the present invention effectively avoids the erroneous termination of the hovering UAV track, reduces the generation of unnecessary new tracks, reduces the number of redundant tracks, greatly optimizes the target tracking results, improves the quality and readability of the UAV radar track, and facilitates the subsequent analysis and judgment of the UAV motion status.

[0012] (3) The present invention proposes a swarm UAV hovering prediction method based on historical track information, which can effectively identify and process UAVs in hovering state, enhance the radar system's ability to judge the hovering of UAV targets, and optimize the data interconnection process by accurately marking and managing the UAV status, ensuring the accuracy of subsequent state estimation.

[0013] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A hovering prediction algorithm flow chart of a swarm UAV hovering prediction method based on historical track information of the present invention; Figure 2 A schematic diagram of the target's real motion trajectory of a swarm UAV hovering prediction method based on historical track information of the present invention; Figure 3 This is a main view of the tracking results of a swarm UAV hovering prediction method based on historical track information of the present invention; Figure 4 This is a top view of the tracking results of a swarm UAV hovering prediction method based on historical track information of the present invention. DETAILED DESCRIPTION

[0015] In order to make the technical solutions, advantages and purposes of the present invention clearer, the technical solutions of the embodiments of the present invention are clearly and completely described below. The described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of this application.

[0016] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0017] like Figure 1 As shown, the present invention provides a method for predicting hovering of a swarm UAV based on historical track information, and the specific steps are as follows: Step S1: Setting speed threshold and radar sampling period threshold ; Step S2: Set the speed and speed threshold of each drone Compare and mark possible hovering drones or remove marks of possible hovering drones; After the state estimation phase at each moment, the speed of each drone at that moment is recorded. , and compare it with the speed threshold For comparison, if If the drone has not been marked as a possible hovering drone, then mark the drone as a possible hovering drone. And if the drone has been marked as a possible hovering drone, remove the mark of the drone.

[0018] Step S3: Obtain the track corresponding to each UAV at the next moment to see whether the associated point track is obtained and mark the UAV that may be hovering, and determine whether to terminate the track; If the track corresponding to a certain UAV at the next moment is not associated with the measurement data, check whether the UAV is marked as a possible hovering UAV. If it is marked as a possible hovering UAV, continue to retain the track for the data interconnection stage and the state estimation stage. If the UAV is not marked as a possible hovering UAV, directly transfer the track to the track termination stage.

[0019] Step S4: Exceeding the radar sampling period threshold The track corresponding to the drone without point track update is terminated.

[0020] If a track corresponds to a drone that is marked as a possible hovering drone and the track has a continuous If there is no update during the radar sampling period, the track will be transferred to the track termination stage.

[0021] Example The target tracking trajectory of this embodiment is obtained by simulating the measured data of multiple UAVs. There are five UAVs in total. Figure 2As shown in the figure, the flight trajectory of each drone is roughly a serpentine trajectory in three-dimensional space. From the figure, we can see that each drone is about 100 meters away from the radar. The flight speed is The first UAV hovered at the 12th sampling moment, the second UAV hovered at the 24th sampling moment, the fourth UAV hovered at the 32nd sampling moment, and the fifth UAV hovered at the 23rd sampling moment. Each UAV had obvious turning maneuvers at multiple radar sampling moments. The specific parameters used in the experiment are shown in Table 1.

[0022] Table 1 Specific parameters used in the experiment

[0023] Step 1: Set the speed threshold is 1 and radar sampling period threshold is 5.

[0024] Step 2: After the state estimation phase at each moment, record the speed of each drone at that moment , and compare it with the speed threshold For comparison, if If the drone has not been marked as a possible hovering drone, then mark the drone as a possible hovering drone. And if the drone has been marked as a possible hovering drone, remove the mark of the drone.

[0025] Step 3: After completing the second step, proceed to the data association stage and state estimation stage at the next moment. If the track corresponding to a certain drone at the next moment is not associated with the measurement data, first check whether this drone is marked as a possible hovering drone. If it is marked as a possible hovering drone, the track will not be transferred to the track termination stage first, and the track will continue to be retained for data interconnection and state estimation. If this drone is not marked as a possible hovering drone, the track will be directly transferred to the track termination stage.

[0026] Step 4: If a track corresponds to a drone that is marked as a possible hovering drone and the track has a continuous If there is no update during a radar sampling period, the track will also be transferred to the track termination stage.

[0027] like Figure 3-4The tracking results are shown in Table 2. The root mean square error (RMSE) of the hover prediction algorithm is 41.0921, and the multi-target tracking accuracy (MOTA) is 1. RMSE reflects the degree of deviation between the tracking result and the true value. A lower RMSE value indicates that the algorithm can make the radar tracking result closer to the real motion trajectory of the drone. A MOTA of 1 indicates that the algorithm has excellent detection, positioning and tracking effects on the target during multi-target tracking, with almost no missed detection and false detection, which further proves the effectiveness of the algorithm in improving radar tracking accuracy. At the same time, the running time of the algorithm is 0.4751 seconds, which means that it can complete the tracking calculation of the swarm drone in a short time and meet the real-time requirements. In actual application scenarios, the fast tracking response can timely grasp the dynamics of the drone, provide sufficient time for subsequent decision-making and countermeasures, and improve the ability and efficiency of responding to the threat of swarm drones.

[0028] Table 2 Performance indicators of hover prediction algorithm

[0029] Therefore, the present invention provides a swarm UAV hovering prediction method based on historical track information, which effectively solves the problems of track fragmentation and increased redundant tracks in traditional track termination algorithms when facing hovering UAVs, improves the accuracy and continuity of radar tracking of swarm UAVs, reduces redundant tracks, and realizes efficient real-time tracking.

[0030] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for predicting hovering of a swarm of drones based on historical track information, characterized in that: The specific steps are as follows: Step S1: Setting speed threshold and radar sampling period threshold ; Step S2: Set the speed and speed threshold of each drone Compare and mark possible hovering drones or remove marks of possible hovering drones; Step S3: Obtain the track corresponding to each UAV at the next moment to see whether the associated point track is obtained and mark the UAV that may be hovering, and determine whether to terminate the track; Step S4: Exceeding the radar sampling period threshold The track corresponding to the drone without point track update is terminated.

2. The method for predicting hovering of a swarm of drones based on historical track information according to claim 1, characterized in that: In step S2, after the state estimation phase at each moment is completed, the speed of each drone at that moment is recorded. , and compare it with the speed threshold For comparison, if and the drone has not been marked as a possible hovering drone, then mark the drone as a possible hovering drone; if And if the drone has been marked as a possible hovering drone, remove the mark of the drone.

3. The method for predicting hovering of a swarm of drones based on historical track information according to claim 2, characterized in that: In step S3, if the track corresponding to a certain UAV at the next moment is not associated with the measurement data, check whether this UAV is marked as a possible hovering UAV. If it is marked as a possible hovering UAV, continue to retain the track for data interconnection and state estimation; if this UAV is not marked as a possible hovering UAV, directly transfer the track to the track termination stage.

4. The method for predicting hovering of a swarm of drones based on historical track information according to claim 3 is characterized in that: In step S4, if a UAV corresponding to a track is marked as a possible hovering UAV and the track has continuous If there is no update during the radar sampling period, the track will be transferred to the track termination stage.

Images