A Cluster Tracking Control Method Based on Maneuvering Target Trajectory Prediction

Abstract

This invention relates to a swarm tracking control method based on maneuvering target trajectory prediction, belonging to the field of unmanned aerial vehicle (UAV) control technology. The swarm tracking control method of this invention predicts the current pose information based on historical data of the target's pose, and then predicts the target's future position. Subsequently, based on the predicted target position, a comprehensive tracking scheme is designed, including "swarm center tracking the predicted target position," "UAVs maintaining distance from the swarm center," and "collision avoidance between UAVs." Corresponding speed control commands are designed to achieve better cooperative tracking performance under maneuvering target scenarios, enabling the swarm to respond quickly to target maneuvers, optimizing target tracking performance, maximizing the swarm's detection of the target, and avoiding collisions between UAVs. This solves the technical problems in existing technologies where the swarm easily loses the target position, causing complete failure of the tracking task, and is prone to collisions.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) control technology, specifically relating to a swarm tracking control method based on the prediction of maneuvering target trajectories. Background Technology

[0002] Cooperative target tracking control in unmanned swarms is a fundamental capability for unmanned swarms to perform tasks and a crucial technology in unmanned swarm cooperative control. It requires unmanned swarms to leverage their numbers and formation advantages to maintain good tracking performance on moving targets. Current research methods mainly include formation tracking based on virtual leaders and formation tracking based on heterogeneous roles.

[0003] Current solutions to the unmanned swarm cooperative target tracking problem mostly transform it into a formation control problem. This involves designing the formation shape for target tracking based on the target's position, followed by a control scheme to achieve the control objective. Generally, this approach assumes the target's motion is smooth, rarely involving high acceleration or high angular velocity maneuvers. The main problem with this method is its poor tracking performance in scenarios with target maneuvering. This is primarily because formation control tracking schemes based on target position have a lag effect on changes in the target's motion state. When the target's motion is stable, the formation error is small, and the swarm tracking effect is good. However, when the target's motion state changes rapidly, i.e., when the target maneuvers, the formation control-based target tracking method struggles to respond quickly. This can cause the target to rapidly escape the swarm's encirclement, or even lead to the swarm losing the target's position, resulting in complete tracking mission failure. Summary of the Invention

[0004] In view of the above problems, this invention provides a swarm tracking control method based on maneuvering target trajectory prediction. The method of this invention obtains the target's average acceleration and average angular velocity based on historical data of the target's pose, and uses this to predict the target's position after a certain time. Then, based on the predicted target position, a comprehensive tracking scheme is designed, including "swarm center tracking the predicted target position," "UAVs maintaining distance from the swarm center," and "collision avoidance between UAVs." Corresponding speed control commands are designed to achieve better cooperative tracking performance under maneuvering target scenarios, enabling the swarm to respond quickly to target maneuvers, optimizing target tracking performance, maximizing the swarm's ability to detect the target, and avoiding collisions between UAVs. This solves the technical problems in existing technologies where the swarm easily loses the target position, causing complete failure of the tracking task, and is prone to collisions.

[0005] This invention provides a cluster tracking control method based on maneuvering target trajectory prediction, comprising the following specific steps:

[0006] S1. Predict the target's current pose information based on the target's historical pose information; predict the target's future pose information based on the predicted value of the target's current pose information; obtain the target's predicted position vector based on the predicted value of the target's future pose information;

[0007] S2. Obtain the coordinate center of the cluster based on the current position vector of each UAV in the cluster; obtain the velocity command component of the cluster's coordinate center to track the predicted position of the target based on the coordinate center of the cluster and the predicted value of the target position vector.

[0008] S3. Obtain the velocity command component that maintains the desired distance between the drone and the cluster's coordinate center;

[0009] S4. Obtain the collision speed commands generated by any two drones to avoid collision; obtain the collision avoidance speed command components between drones based on the collision speed commands;

[0010] S5. Based on the speed command components obtained in steps S2, S3, and S4, combine them to form the speed command of the UAV.

[0011] Optionally, in step S1, the target's historical pose information includes estimated values ​​of linear acceleration and heading angular velocity, expressed as:

[0012]

[0013]

[0014] Among them, a 0e (t-τ l +iT min ), ω 0e (t-τ l +iT min ) represent the target at t-τ l +iT min Estimates of linear acceleration and angular velocity at time t; v0(t-τ) l +(i+1)T min v0(t-τ) l +iT min ) represent the target at t-τ l +(i+1)T min Time, t-τ l +iT min The linear velocity at time t; θ0(t-τ) l +(i+1)T min ), θ0(t-τ l +iT min ) represent the target at t-τ l +(i+1)T min Time, t-τ l+iT min The heading angle at time; t represents the current time; i represents the current time of the i-th sampling; i+1 represents the current time of the (i+1)-th sampling; T min Indicates the sampling time; k l This indicates the number of times the historical location data of the target was sampled. τ l The duration of historical location data collection for the target.

[0015] Optionally, in step S1, the target's current pose information includes the estimated target linear acceleration and the estimated heading angular velocity at the current time t, expressed as:

[0016]

[0017]

[0018] Among them, a 0em (t) represents the estimated linear acceleration of the target at the current time t; a 0e (t-τ l +(i+1)T min ) indicates that the target is at t-τ l +(i+1)T min Linear acceleration at time ω; 0em (t) represents the estimated heading angular velocity of the target at the current time t; ω 0e (t-τ l +(i+1)T min ) indicates that the target is at t-τ l +(i+1)T min The heading angular velocity at a given moment.

[0019] Optionally, in step S1, the target's future pose information includes the target's future position information within the future time period [t, t+τ]. p The estimated values ​​of linear velocity and angular velocity of [ ] are expressed as follows:

[0020] v 0p (t+jT min ) = v 0p (t+(j-1)T min )+a 0em (t)T min j = 1, 2, ..., k p

[0021] θ 0p (t+jT min )=θ 0p (t+(j-1)T min )+ω 0em (t)T minj = 1, 2, ..., k p

[0022] Among them, v 0p (t+jT min ) indicates that the target is at t+jT min Predicted linear velocity at time v; 0p (t+(j-1)T min ) indicates that the target is at t+(j-1)T min Predicted linear velocity at time θ 0p (t+jT min ) indicates that the target is at t+jT min Predicted angular velocity at time θ 0p (t+(j-1)T min ) indicates that the target is at t+(j-1)T min The predicted heading angular velocity at time point; kp represents the number of times the target linear velocity and heading angular velocity are predicted. τ p The target historical location data collection duration and the target prediction time period; j indicates that the current prediction time is the j-th sampling.

[0023] Optionally, in step S1, the expression for the target predicted location vector is:

[0024]

[0025] in, Indicates the target at t+τ p Predicted position vector at time; Indicates the target at t+τ p The predicted x-axis position at time 10:00; Indicates the target at t+τ p Predicted y-axis position at time p; 0x (t) represents the x-axis position of the target at time t; p 0y (t) represents the y-axis position of the target at time t; v 0x (t) represents the target's x-axis velocity at time t; v 0y (t) represents the y-axis velocity of the target at time t.

[0026] Optionally, in step S2, the position vector of the nth UAV in the cluster at the current time t. The expression is:

[0027]

[0028] Where, p nx (t) represents the x-axis position of the nth drone at the current time t; p ny(t) represents the y-axis position of the nth drone at the current time t; N represents the total number of drones in the cluster.

[0029] Optionally, in step S2, the coordinate center of the cluster at the current time t The expression is:

[0030]

[0031] Optionally, in step S2, the velocity command component of the nth UAV used to realize the function of tracking the predicted position of the target by the cluster center is obtained based on the difference between the coordinate center of the cluster and the predicted position vector of the target at the current time t. The expression is:

[0032]

[0033] Optionally, the specific steps of step S3 are as follows:

[0034] S31. Obtain the vector pointing from the position of the nth drone to the center of the cluster. The expression is:

[0035]

[0036] S32. Based on the difference between the distance between the UAV and the cluster's coordinate center and the desired distance, obtain the velocity command component for the i-th UAV to maintain distance from the cluster center. The expression is:

[0037]

[0038] Where, d aim This represents the desired distance.

[0039] Optionally, the specific steps of step S4 are as follows:

[0040] S41. Obtain the vector pointing from the m-th drone to the n-th drone. The expression is:

[0041]

[0042] in, Let m represent the position vector of the m-th UAV at the current time t, where m = 1, 2, ..., N;

[0043] S42. Obtain the collision avoidance speed command generated by the nth drone to avoid colliding with the mth drone. The expression is:

[0044]

[0045] Where ε is a minimal positive number; d safe Indicates the safe distance between drones;

[0046] S43. Considering the collision avoidance speed commands of the nth drone and all other drones in the cluster, obtain the drone-to-drone collision avoidance speed command component of the nth drone. The expression is:

[0047]

[0048] Compared with existing technologies, the present invention has at least the following beneficial effects: The method of the present invention pre-adjusts the target's maneuvering behavior by using speed commands from the cluster center to track the predicted target position, improving robustness against target maneuvering behavior; ensures good detection and encirclement of the maneuvering target by using speed commands to maintain distance between the UAV and the cluster center; and ensures target tracking safety by using speed commands to avoid collisions between UAVs to maintain a distance greater than a safe distance between UAVs. The speed commands corresponding to these three cooperative behaviors, under specific weights, constitute a cooperative target tracking scheme, achieving the control objective of safely and efficiently tracking maneuvering targets. Compared with cooperative target tracking methods based on virtual leaders, the method of the present invention has better responsiveness to target maneuvering and better tracking performance in situations such as rapid target acceleration; compared with cooperative target tracking methods based on heterogeneous role allocation, the method of the present invention does not rely on a fixed number of UAVs in the unmanned cluster, making the requirement for the number of UAVs in the cluster relatively flexible. Overall, the present invention addresses the needs of cooperative tracking of maneuvering targets and the limitations of existing cooperative methods in dealing with target maneuvering. The proposed cooperative tracking method has the potential for initial application in maneuvering target tracking, and the effectiveness of the tracking method in scenarios of target acceleration and rotation has been tested through numerical simulation. Attached Figure Description

[0049] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of the invention.

[0050] Figure 1 Speed ​​command of the present invention Diagram illustrating the control effect;

[0051] Figure 2 Speed ​​command of the present invention Diagram illustrating the control effect;

[0052] Figure 3 Speed ​​command of the present invention Diagram illustrating the control effect;

[0053] Figure 4 This is a schematic diagram of the coverage area of ​​the drone camera of the present invention;

[0054] Figures 5(a)-(b) show the simulation results of the motion trajectories of the unmanned swarm and the target under Strategy 1 and Strategy 2 of the present invention;

[0055] Figures 6(a)-(b) show the simulation results of unmanned cluster tracking of targets at various time points for Strategy 1 and Strategy 2 of the present invention;

[0056] Figure 7 These are the simulation results of the distance between the cluster center and the target for Strategy 1 and Strategy 2 of this invention;

[0057] Figure 8 The simulation results show the number of UAVs that detected the target in Strategy 1 and Strategy 2 of this invention.

[0058] Figures 9(a)-(b) show the simulation results of the distance between the UAV and the cluster center for Strategy 1 and Strategy 2 of the present invention;

[0059] Figures 10(a)-(b) show the simulation results of the distance between UAVs for Strategy 1 and Strategy 2 of the present invention;

[0060] Figures 11(a)-(c) show the simulation results of the kinematic characteristics of UAVs under Strategy 1 and Strategy 2 of the present invention. Detailed Implementation

[0061] To better understand the above-described objectives, features, and advantages of the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other. Furthermore, the present invention can be implemented in other ways different from those described herein; therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0062] A specific embodiment of the present invention, such as Figure 1-1 1. A cluster tracking control method based on maneuvering target trajectory prediction is disclosed, including the following specific steps:

[0063] S1. Obtain the target's predicted location vector;

[0064] The target prediction location vector of the present invention It is based on the target's pose information at time t, [t-τ] l The target obtained from the historical pose information of the time period [t] is in the range of t+τ. p The predicted position vector at time t. The unmanned swarm will be based on the predicted position vector at time t. The design incorporates speed control commands to coordinate with the target's maneuvering motion, thereby proactively responding to the target's maneuvering.

[0065] S11. Obtain the target within the historical time period [t-τ] lThe estimated values ​​of linear acceleration and directional angular velocity at each time point in [t] are expressed as follows:

[0066]

[0067]

[0068] Among them, a 0e (t-τ l +iT min ), ω 0e (t-τ l +iT min ) represent the target at t-τ l +iT min Estimates of linear acceleration and angular velocity at time t; v0(t-τ) l +(i+1)T min v0(t-τ) l +iT min ) represent the target at t-τ l +(i+1)T min Time, t-τ l +iT min The linear velocity at time t; θ0(t-τ) l +(i+1)T min ), θ0(t-τ l +iT min ) represent the target at t-τ l +(i+1)T min Time, t-τ l +iT min The heading angle at time; t represents the current time; i represents the current time of the i-th sampling; i+1 represents the current time of the (i+1)-th sampling; T min Indicates the sampling time; k l This indicates the number of times the historical location data of the target was sampled. τ l The duration of historical location data collection for the target;

[0069] The estimated values ​​of the target line acceleration and heading angular velocity at the current time t are obtained using the following expressions:

[0070]

[0071]

[0072] Among them, a 0em (t) represents the estimated linear acceleration of the target at the current time t; a 0e (t-τ l +(i+1)T min) indicates that the target is at t-τ l +(i+1)T min Linear acceleration at time ω; 0em (t) represents the estimated heading angular velocity of the target at the current time t; ω 0e (t-τ l +(i+1)T min ) indicates that the target is at t-τ l +(i+1)T min The heading angular velocity at a given moment.

[0073] S12, Predict the target in the future time period [t, t+τ] p The estimated values ​​of linear velocity and angular velocity of [ ] are expressed as follows:

[0074] v 0p (t+jT min ) = v 0p (t+(j-1)T min )+a 0em (t)T min j = 1, 2, ..., k p

[0075] θ 0p (t+jT min )=θ 0p (t+(j-1)T min )+ω 0em (t)T min j = 1, 2, ..., k p

[0076] Among them, v 0p (t+jT min ) indicates that the target is at t+jT min Predicted linear velocity at time v; 0p (t+(j-1)T min ) indicates that the target is at t+(j-1)T min Predicted linear velocity at time θ 0p (t+jT min ) indicates that the target is at t+jT min Predicted angular velocity at time θ 0p (t+(j-1)T min ) indicates that the target is at t+(j-1)T min The predicted heading angular velocity at time point; kp represents the number of times the target linear velocity and heading angular velocity are predicted. τ p The target historical location data collection duration and the target prediction time period; j indicates that the current prediction time is the j-th sampling.

[0077] Furthermore, when j=1, it is considered that:

[0078]

[0079] Where v0(t) and θ0(t) are the linear velocity and heading angular velocity of the target at time t, respectively.

[0080] S13. Obtain the target predicted position vector The expression is:

[0081]

[0082] in, Indicates the target at t+τ p Predicted position vector at time; Indicates the target at t+τ p The predicted x-axis position at time 10:00; Indicates the target at t+τ p Predicted y-axis position at time p; 0x (t) represents the x-axis position of the target at time t; p 0y (t) represents the y-axis position of the target at time t; v 0x (t) represents the target's x-axis velocity at time t; v 0y (t) represents the y-axis velocity of the target at time t.

[0083] S2. Obtain the velocity command component of the target's predicted position at the coordinate center of the cluster.

[0084] The velocity command component of the cluster's coordinate center tracking target predicted position behavior in this invention As the core of the tracking control strategy, it enables the cluster's center of mass to track the predicted location of the target. See control effect Figure 1 .

[0085] S21. Obtain the position vector of the nth UAV in the cluster at the current time t, expressed as:

[0086]

[0087] Where, p nx (t) represents the x-axis position of the nth drone at the current time t; p ny (t) represents the y-axis position of the nth drone at the current time t; N represents the total number of drones in the cluster;

[0088] S22. Obtain the coordinate center of the cluster (the coordinate center of N drones) at the current time t:

[0089]

[0090] S23. Based on the difference between the coordinate center of the cluster and the predicted position vector of the target at the current time t, obtain the velocity command component of the nth UAV used to realize the cluster center tracking the predicted position of the target. The expression is:

[0091]

[0092] S3. Obtain the velocity command component that maintains the desired distance between the drone and the cluster's coordinate center;

[0093] This invention relates to the velocity command component that allows the UAV to maintain distance from the cluster center while accurately tracking the predicted target position. Used to ensure that the distance between the UAV and the target is within the selected desired distance d. aim Within its neighborhood. See control effect Figure 2 .

[0094] S31. Obtain the vector pointing from the position of the nth drone to the center of the cluster, expressed as:

[0095]

[0096] S32. Based on the distance between the coordinate centers of the UAV and the cluster and the desired distance d aim The difference is used to obtain the velocity command component of the i-th UAV maintaining distance from the cluster center, expressed as:

[0097]

[0098] Where, d aim This represents the desired distance, which ensures that the target is within the detection range of the drone's sensors, and that the drone maintains a large distance from the target to accommodate its maneuvering.

[0099] S4. Obtain the speed command components for collision avoidance between drones;

[0100] Based on the velocity command components defined in S2 and S3, this invention enables the cluster to achieve good target tracking. Furthermore, it provides velocity command components for collision avoidance between UAVs. Used to avoid collisions between drones and ensure the safety of drone flight. See control effect Figure 3 .

[0101] S41. Obtain the vector pointing from the m-th UAV to the n-th UAV, expressed as:

[0102]

[0103] in, Let m represent the position vector of the m-th UAV at the current time t, where m = 1, 2, ..., N;

[0104] S42. Obtain the collision avoidance speed command generated by the nth drone to avoid colliding with the mth drone. The expression is:

[0105]

[0106] Where ε is a minimal positive number; d safe This indicates the safe distance between drones.

[0107] S43. Considering the collision avoidance speed commands of the nth drone and all other drones in the cluster, obtain the drone-to-drone collision avoidance speed command component of the nth drone, expressed as:

[0108]

[0109] S5. Obtain the combined speed command from the drone.

[0110] Based on the speed command components obtained from S2, S3, and S4, the speed command of the UAV is formed by combining them. as follows:

[0111]

[0112] Where k1, k2, and k3 are positive real number weight coefficients.

[0113] To illustrate the effectiveness of the method proposed in this invention, the following detailed description of the above technical solution is provided through a specific embodiment. The specific implementation steps are as follows:

[0114] 1. Modeling the motion of fixed-wing unmanned swarms and targets:

[0115] Suppose there are 6 fixed-wing UAVs in the unmanned swarm performing a cooperative target tracking task. The kinematic model of the UAVs is as follows:

[0116]

[0117] in, p represents the position vector of drone i. ix p represents the x-direction position of drone i; iy This indicates the y-direction position of drone i; Represents the heading angle and linear velocity of UAV i, [ω i ,a i ] TLet ω be the angular velocity and linear acceleration, which are the control inputs to the UAV's state vector i. The fixed-wing UAV is subject to the following constraints: maximum flight speed, maximum angular velocity, and maximum acceleration.

[0118] v i ∈[v min ,v max ]

[0119] |ω i |≤ω max

[0120] |a i |≤a max

[0121] Among them, v min ,v max These represent the minimum and maximum flight speeds of the drone, respectively; ω max Indicates the maximum angular velocity of the drone; a max This indicates the maximum acceleration of the drone.

[0122] In addition, assuming that the drone perceives the target using camera detection, and assuming that the area covered by the drone's camera is as follows: Figure 4 As shown.

[0123] 2. Brief Description of the Case Comparison Strategy

[0124] This implementation example compares two technical strategies to verify the effectiveness of the technical solution proposed in this invention. The only difference between the two strategies is whether or not the target's position is predicted before tracking.

[0125] Strategy 1 modifies the speed command for tracking the predicted position of the target at the cluster center in the technical solution proposed in this invention to:

[0126]

[0127] That is, the target's motion is not predicted, allowing the cluster center to track the target's current position. Furthermore, the speed commands for maintaining distance between the drone and the cluster center, and the speed commands for collision avoidance between drones, are the same as those proposed in this invention; that is, the final speed command for Strategy 1 is...

[0128]

[0129] Strategy 2 uses the same technical solution as proposed in this invention, that is, the final speed command of Strategy 1 is...

[0130]

[0131] The two strategies use the same weight coefficients for k1, k2, and k3.

[0132] 3. Simulation initial conditions and simulation parameter settings

[0133] During the simulation, Strategy 1 and Strategy 2 used the same initial conditions, simulation parameters, and target motion patterns. The kinematic constraints on the UAV and the tracked target are shown in Table 1. The maximum speed of the tracked target is greater than that of the UAV, meaning the target speed can be greater than the UAV's maximum speed, which increases the tracking difficulty. However, the maximum acceleration of the UAV is slightly greater than that of the target, giving it slightly greater maneuverability.

[0134] Table 1. Capability Constraints of UAVs and Targets

[0135] parameter drone constraints Tracked target constraints unit <![CDATA[Velocity v i > [25,90] [25,100] m / s <![CDATA[Angular velocity ω i > [-10,10] [-10,10] ° / s <![CDATA[Acceleration a i > [-11,11] [-10,10] <![CDATA[m / s 2 ]]>

[0136] The initial states of the UAV and the target are shown in Table 2.

[0137] Table 2 Initial State of the UAV and Target

[0138]

[0139]

[0140] The target's movement is a turning acceleration motion, with a movement time of 9 seconds, and the movement pattern is shown in Table 3.

[0141] Table 3 Target Motion Patterns

[0142]

[0143] See Table 4 for speed command parameter selection, where d is selected. aim =140m, this distance ensures that the drone can maintain good detection of the target and prevent the target from being lost.

[0144] Table 4 Speed ​​Command Parameter Selection

[0145] parameter Numerical values ​​(units omitted) <![CDATA[k1]]> 2 <![CDATA[k2]]> 1.2 <![CDATA[k3]]> 2000 <![CDATA[d aim ]]> 140 <![CDATA[d safe ]]> 100

[0146] 4. Based on the technical solution in the invention, the program is designed using MATLAB for UAVs. The execution flow of the proposed swarm tracking control method based on maneuvering target trajectory prediction for strategy 1 or strategy 2 is as follows:

[0147]

[0148] 5. Output and analyze simulation results

[0149] Figure 5 shows the motion trajectories of the unmanned swarms and the target during the simulation, while Figure 6 shows the tracking status of the unmanned swarms at various times in both strategies 1 and 2. The distances between the swarm centers and the target in both strategies 1 and 2 are shown in Figure 6. Figure 7 As shown, the number of drones detecting the target in the strategy 1 and strategy 2 clusters is as follows: Figure 8 As shown in Figure 9, the distance between the unmanned swarm drones and the swarm center in Strategy 1 and Strategy 2 is shown in Figure 10, and the kinematic characteristics of the drones in Strategy 1 and Strategy 2 are shown in Figure 11.

[0150] Specifically, in the set of figures in the simulation results, the first figure in Figures 5, 6, 9, and 10 shows the simulation results for Strategy 1, and the second figure shows the simulation results for Strategy 2. In Figure 11, the first figure shows the velocity curve, the second figure shows the acceleration curve, and the third figure shows the angular velocity curve.

[0151] Figures 5 and 6 Figure 7 The simulation results reflect the target tracking performance of the cluster during the motion of the target under strategies 1 and 2. The simulation results show that strategy 1 has poor target tracking performance when the target is accelerating, and the distance between the cluster center and the target is large. In contrast, strategy 2 optimizes the target tracking performance after predicting the target position, and the cluster center maintains a smaller distance from the target. This shows that the tracking behavior of the cluster center tracking the predicted position of the target is effective for tracking moving targets.

[0152] Figure 8 The simulation results reflect the target detection effectiveness of the drone swarm in Strategy 1 and Strategy 2. The lower the number of drones, the higher the probability that the target will escape the tracking of multiple drones. The simulation results show that the number of drones detected by Strategy 1 decreases significantly after 5 seconds due to the target's maneuvering (less than 2 drones), while Strategy 2 maintains a better detection effect by detecting more than 5 drones due to the drones' movement.

[0153] Figure 9 illustrates the distance between the UAV and the swarm center under strategies 1 and 2. Simulation results show that although both strategies follow the same form for the speed control command regarding the distance between the UAV and the swarm center, in strategy 1, when the swarm center cannot effectively track the target, the distance between the UAV and the swarm center cannot be maintained well at the set distance d. aim Nearby, while strategy 2 can maintain the distance between the drone and the cluster center at a set d. aim nearby.

[0154] Figure 10 illustrates the distances between the unmanned aerial vehicles (UAVs) in the swarm under strategies 1 and 2. Simulation results show that both strategies 1 and 2 can ensure that the distance between UAVs is greater than the safe distance, thus preventing collisions.

[0155] Figure 11 shows the velocity, acceleration, and angular velocity curves of each UAV in Strategy 1 and Strategy 2. Simulation results show that Strategy 2 has a faster response speed to the target's acceleration and turning compared to Strategy 1.

[0156] The simulation results above verify that the cluster tracking control method based on maneuvering target trajectory prediction proposed in this invention can be applied to tracking maneuvering targets and achieve target tracking effect when the target is accelerating and turning.

[0157] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A swarm tracking control method based on maneuvering target trajectory prediction, characterized in that, The specific steps include the following: S1. Predict the target's current pose information based on the target's historical pose information; predict the target's future pose information based on the predicted value of the target's current pose information; obtain the target's predicted position vector based on the predicted value of the target's future pose information; S2. Obtain the coordinate center of the cluster based on the current position vectors of each UAV in the cluster; based on the coordinate center of the cluster and the predicted value of the target position vector, obtain the velocity command component of the cluster's coordinate center to track the predicted position of the target. The position vector of the nth drone in the cluster at time t. The expression is: Where, p nx (t) represents the x-axis position of the nth drone at the current time t; p ny (t) represents the y-axis position of the nth drone at the current time t; N represents the total number of drones in the cluster; The coordinate center of the cluster at the current time t The expression is: The velocity command component for the nth UAV, used to achieve the function of tracking the predicted position of the target by the cluster center, is obtained based on the difference between the coordinate center of the cluster and the predicted position vector of the target at the current time t. The expression is: in, Indicates the target at t+τ p Predicted position vector value at time τ; p Predict the duration for the target; S3. Obtain the velocity command components that allow the UAV and the cluster to maintain the desired distance between their coordinate centers. The specific steps are as follows: S31. Obtain the vector pointing from the position of the nth drone to the center of the cluster. The expression is: S32. Based on the difference between the distance between the UAV and the cluster's coordinate center and the desired distance, obtain the velocity command component for the nth UAV to maintain distance from the cluster center. The expression is: Where, d aim Indicates the expected distance; S4. Obtain the collision speed commands generated by any two drones to avoid collision; based on the collision speed commands, obtain the collision avoidance speed command components between the drones. The specific steps are as follows: S41. Obtain the vector pointing from the m-th drone to the n-th drone. The expression is: in, Let m represent the position vector of the m-th UAV at the current time t, where m = 1, 2, ..., N; S42. Obtain the collision avoidance speed command generated by the nth drone to avoid colliding with the mth drone. The expression is: Where ε is a minimal positive number; d safe Indicates the safe distance between drones; S43. Considering the collision avoidance speed commands of the nth drone and all other drones in the cluster, obtain the drone-to-drone collision avoidance speed command component of the nth drone. The expression is: S5. Based on the speed command components obtained in steps S2, S3, and S4, combine them to form the speed command of the UAV. The expression is: Where k1, k2, and k3 are positive real number weight coefficients.

2. The cluster tracking control method according to claim 1, characterized in that, In step S1, the target's historical pose information includes estimated values ​​of linear acceleration and heading angular velocity, expressed as: Among them, a 0e (t-τ l +iT min ), ω 0e (t-τ l +iT min ) represent the target at t-τ l +iT min Estimates of linear acceleration and angular velocity at time t; v0(t-τ) l +(i+1)T min v0(t-τ) l +iT min ) represent the target at t-τ l +(i+1)T min Time, t-τ l +iT min The linear velocity at time t; θ0(t-τ) l +(i+1)T min ), θ0(t-τ l +iT min ) represent the target at t-τ l +(i+1)T min Time, t-τ l +iT min The heading angle at time; t represents the current time; i represents the current time of the i-th sampling; i+1 represents the current time of the (i+1)-th sampling; T min Indicates the sampling time; k l This indicates the number of times the historical location data of the target was sampled. τ l The duration of historical location data collection for the target.

3. The cluster tracking control method according to claim 2, characterized in that, In step S1, the target's current pose information includes the estimated target linear acceleration and the estimated heading angular velocity at the current time t, expressed as: Among them, a 0em (t) represents the estimated linear acceleration of the target at the current time t; a 0e (t-τ l +(i+1)T min ) indicates that the target is at t-τ l +(i+1)T min Linear acceleration at time ω; 0em (t) represents the estimated heading angular velocity of the target at the current time t; ω 0e (t-τ l +(i+1)T min ) indicates that the target is at t-τ l +(i+1)T min The angular velocity of heading at any given moment.

4. The cluster tracking control method according to claim 3, characterized in that, In step S1, the target's future pose information includes the target's future position within the time interval [t, t+τ]. p The estimated values ​​of linear velocity and heading angle of [ ] are expressed as follows: v 0p (t+jT min )=v 0p (t+(j-1)T min )+a 0em (t)T min ,j=1,2,…,k p θ 0p (t+jT min )=θ 0p (t+(j-1)T min )+ω 0em (t)T min ,j=1,2,…,k p Among them, v 0p (t+jT min ) indicates that the target is at t+jT min Predicted linear velocity at time v; 0p (t+(j-1)T min ) indicates that the target is at t+(j-1)T min Predicted linear velocity at time θ 0p (t+jT min ) indicates that the target is at t+jT min Predicted heading angle at time θ 0p (t+(j-1)T min ) indicates that the target is at t+(j-1)T min Predicted heading angle at time; k p This indicates the number of times the target's linear velocity and heading angle are predicted. j represents the current prediction time as the j-th sample.

5. The cluster tracking control method according to claim 4, characterized in that, In step S1, the expression for the target predicted position vector is: in, Indicates the target at t+τ p The predicted x-axis position at time 10:00; Indicates the target at t+τ p Predicted y-axis position at time p; 0x (t) represents the x-axis position of the target at time t; p 0y (t) represents the y-axis position of the target at time t; v 0x (t) represents the target's x-axis velocity at time t; v 0y (t) represents the y-axis velocity of the target at time t.

Images