A dynamic task allocation and cooperative guidance control system and method for multi-unmanned aerial vehicle cooperative containment of a maneuvering target

Abstract

A kind of dynamic task allocation and cooperative guidance control system and method for multi-unmanned aerial vehicle cooperative containment of mobile target.The application is based on the dynamic containment formation generation method of target real-time heading angle, so that the relative position of containment position is always kept unchanged in the direction of target motion, dynamic follow-up containment is realized;Hungarian algorithm frame-by-frame global optimal task allocation strategy eliminates trajectory intersection and minimizes total maneuvering cost;Residual time consistency cooperative guidance strategy takes the time required by the farthest unmanned aerial vehicle as the global synchronization reference, fuses target feedforward speed, ensures that all interceptors reach their positions at the same time and continuously track the mobile target;Lightweight anti-collision method embeds artificial potential repulsion as a speed vector correction term into the speed planning layer;The organic combination of the above four technologies, RK4 high-precision dynamics integration and PID speed tracking control forms a "perception-planning-control" three-layer closed-loop architecture, which realizes stable containment under the condition that the target makes continuous spiral maneuvering.

Description

Technical Field

[0001] This invention belongs to the field of multi-UAV cooperative control and guidance technology, specifically relating to a dynamic task allocation and cooperative guidance control method for multi-UAV cooperative containment of mobile targets. Background Technology

[0002] The classic proportional navigation (PN) guidance law is designed for single-aircraft tracking of a single target, and cannot meet the tactical requirements of multiple aircraft arriving simultaneously and coordinating encirclement. When each interceptor sprints independently, the closest drone arrives alone in advance, exposing a gap before the encirclement is formed, and the target can escape through the gap.

[0003] Fixed formation methods predefine the geometric formation of the aircraft group, with each interceptor maintaining a fixed relative position as it flies toward the target area. When the target makes a turning maneuver, the formation orientation does not adjust accordingly, severely degrading the encirclement effect; moreover, it typically uses a one-time static task allocation, which is not adjusted during flight, making it unable to adapt to dynamically changing encirclement scenarios.

[0004] Traditional multi-machine collaborative methods mostly do not consider the collision risk between interceptors in dense encirclement conditions, or require the introduction of independent obstacle avoidance path planning modules, which increases system complexity and computational overhead. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a dynamic task allocation and collaborative guidance control system and method for multi-UAV collaborative containment of maneuvering targets. It constructs a complete three-dimensional collaborative containment system for multiple UAVs, enabling dynamic formation following of targets undergoing continuous and complex maneuvers. The system is deployed within a UAV swarm with minimal computational overhead, ensuring globally optimal task allocation, synchronized arrival, precise speed tracking, and real-time collision avoidance during operation. This guarantees simultaneous encirclement by multiple UAVs and maintains a stable containment posture.

[0006] The technical solution adopted in this invention is:

[0007] A dynamic task allocation and cooperative guidance control system for multi-UAV cooperative containment of mobile targets, including...

[0008] The perception layer is used to acquire the target aircraft's position, speed, and heading angle in real time.

[0009] The planning layer is used to sequentially complete three tactical calculations within each control step: dynamic blocking position generation, optimal task allocation using the Hungarian algorithm, and global synchronization time coordination.

[0010] The control layer converts the desired velocity into an acceleration command, superimposes it with an artificial potential field obstacle avoidance component, and then advances the physical state through a fourth-order Runge-Kutta integrator.

[0011] A dynamic task allocation and cooperative guidance control method for multi-UAV cooperative containment of maneuvering targets includes the following steps:

[0012] S1. Construct a flight dynamics model for the UAV;

[0013] S2. Production containment formation: With the target's current position as the center and the target's heading angle in the horizontal plane as the reference direction, uniformly distributed containment positions are generated around it. The positions are updated in real time according to the target's position and heading.

[0014] S3. Perform frame-by-frame optimal task allocation using the Hungarian algorithm, modeling task allocation as an assignment problem. Construct a cost matrix using the Euclidean distance from each machine to each array position. Call the Hungarian algorithm in real time at each step to solve for the optimal allocation, eliminate trajectory intersections, and minimize the total maneuver cost.

[0015] S4. Remaining Time Consistent Cooperative Guidance: The time required for the UAV furthest from the target position is used as the global synchronization benchmark, and the remaining UAVs actively slow down and wait to ensure that all interceptors arrive at their respective positions at the same time; at the same time, the target feedforward velocity is integrated to achieve continuous tracking of moving targets.

[0016] S5. PID speed tracking control, responsible for connecting the desired velocity vector and acceleration command;

[0017] S6. Artificial potential field collision avoidance: The repulsive force of the artificial potential field is directly superimposed as a velocity vector correction term onto the desired guidance velocity.

[0018] S7. Determine whether the containment was successful.

[0019] Compared with the prior art, the present invention has the following advantages:

[0020] 1. Dynamic adaptive containment capability: The containment formation is dynamically generated based on the target's real-time heading, and the position points always remain in their relative positions to the target's direction of movement. Combined with frame-by-frame optimal task allocation, the system can maintain an effective containment posture against targets moving in any direction, exhibiting strong robustness.

[0021] 2. Time coordination accuracy: The remaining time consistency strategy ensures that all interceptors arrive at their respective positions at the same time, minimizing the target's escape space; feedforward velocity fusion eliminates tracking lag for moving targets, improving the success rate of containment.

[0022] 3. Flight safety assurance: The artificial potential field collision avoidance is embedded in the velocity planning layer, eliminating the need for a separate obstacle avoidance module, minimizing computational load, ensuring high real-time performance, and naturally compatibility with PID speed controllers, guaranteeing zero collisions in densely packed situations.

[0023] 4. High-precision simulation verification: The RK4 integrator, combined with first-order inertial delay modeling, provides high simulation accuracy and strong numerical stability. The simulation results closely resemble the dynamic characteristics of real UAVs, avoiding overly optimistic assessments from idealized simulations.

[0024] 5. High scalability: The system architecture supports flexible expansion of the number of interceptors, and the formation generation and task allocation algorithms can be naturally extended to [other systems]. A scene of aircraft blocking the way. Attached Figure Description

[0025] Figure 1 This is the overall architecture of the UAV collaborative containment in this invention;

[0026] Figure 2 This is the moment when the simulation system in this invention successfully contained the target. Detailed Implementation

[0027] To better understand the purpose, structure, and function of this invention, the invention will be described in further detail below with reference to the accompanying drawings.

[0028] The present invention is applicable to systems that require multiple interceptor drones to coordinate and block mobile targets in scenarios such as multi-rotor / fixed-wing drone swarm confrontation, low-altitude defense, key area protection, and anti-drone swarm warfare.

[0029] This invention provides a dynamic task allocation and cooperative guidance control method for multi-UAV cooperative interception of maneuvering targets. It employs a dynamic interception formation generation method based on the target's real-time heading angle, ensuring the interception positions remain constant relative to the target's direction of motion, achieving dynamic following and interception. A Hungarian algorithm-based frame-by-frame globally optimal task allocation strategy re-solves the optimal allocation for each simulation step, eliminating trajectory intersections and minimizing the total maneuver cost. A remaining time consistency cooperative guidance strategy uses the time required by the furthest UAV as the global synchronization benchmark, integrating the target's feedforward velocity to ensure all interceptors arrive at their respective positions simultaneously and continuously track the moving target. A lightweight collision avoidance method embeds artificial potential field repulsion as a velocity vector correction term into the velocity planning layer, eliminating the need for a separate obstacle avoidance module, minimizing computational complexity, and physically guaranteeing zero collisions. The organic integration of these four technologies with RK4 high-precision dynamic integral and PID velocity tracking control forms a three-layer closed-loop architecture of "perception-planning-control," achieving stable interception even under conditions of continuous spiral maneuvering of the target.

[0030] The system and method of this invention adopt a three-layer architecture of "sensing-planning-control".

[0031] Perception layer: Real-time acquisition of the target aircraft's position, velocity, and heading angle;

[0032] Planning layer: Within each control step, three tactical calculations are completed sequentially: dynamic blocking position generation, optimal task allocation using the Hungarian algorithm, and global synchronization time coordination (remaining time consistency cooperative guidance strategy).

[0033] In discrete time This indicates the control step size, where the step size is... .

[0034] Control layer: Each interceptor converts the desired velocity into an acceleration command through a PID speed loop, superimposes the obstacle avoidance component of the artificial potential field, and then advances the physical state through a fourth-order Runge-Kutta (RK4) integrator.

[0035] in:

[0036] Dynamic containment formation generation based on target heading: With the target's current position as the center and the target's heading angle in the horizontal plane as the reference direction, containment positions are generated evenly distributed around it. The positions are updated in real time with the target's position and heading, realizing dynamic following containment.

[0037] The Hungarian algorithm for frame-by-frame global optimal task allocation models task allocation as an assignment problem. It constructs a cost matrix based on the Euclidean distance from each machine to each position and calls the Hungarian algorithm in real time at each step to solve for the optimal allocation, eliminate trajectory intersections, and minimize the total maneuver cost.

[0038] Remaining time consistency cooperative guidance strategy: The time required for the UAV farthest from the target position is used as the global synchronization benchmark, and the remaining UAVs actively slow down and wait to ensure that all interceptors arrive at their respective positions at the same time; at the same time, the target feedforward velocity is integrated to achieve continuous tracking of moving targets.

[0039] Artificial potential field collision avoidance embedded in velocity planning layer: The repulsive force of artificial potential field is directly superimposed on the desired velocity of the guidance as a velocity vector correction term. No independent obstacle avoidance module is required, the amount of computation is extremely small, and the repulsive force adopts the reciprocal difference model to physically guarantee zero collision.

[0040] High-precision RK4 dynamics integration and first-order inertial delay modeling: A fourth-order Runge-Kutta integrator is used to introduce a first-order inertial element to simulate the response delay of the dynamic system, so that the simulation results are close to the dynamic characteristics of real UAVs.

[0041] The method of the present invention, such as Figure 1 As shown, it includes the following steps:

[0042] S1. Constructing the UAV flight dynamics model

[0043] Maintain a 9-dimensional state vector for each drone. :

[0044] (1)

[0045] in These are the position, velocity, and acceleration vectors, respectively.

[0046] The continuous-time dynamic equations are defined as follows:

[0047] (2)

[0048] in For command acceleration, Let be the time constant of the dynamic system. This is the derivative of position with respect to time (i.e., velocity). This is the derivative of velocity with respect to time (i.e., acceleration).

[0049] This describes a first-order inertial element, where the actual acceleration is expressed as a time constant. The exponent converges to the command value, which is the physical delay characteristic of the UAV power system.

[0050] To prevent the drone from going out of control, an overload limit is applied to the command acceleration:

[0051] (3)

[0052] in For the maximum permissible overload, This is the acceleration due to gravity.

[0053] Numerical integration is performed using the fourth-order Runge-Kutta method (RK4). The derivative function is sampled four times within each step and then weighted and averaged to reduce the global truncation error to a minimum. This significantly improves the numerical stability of long-term simulations. The four slopes of RK4 are calculated as follows:

[0054] (4)

[0055] The state update formula is:

[0056] (5)

[0057] in For the first The update status of the step.

[0058] S2. Generate encirclement formation

[0059] The algorithm is as follows: using the target's current position With the formation center as the reference point, and the heading angle within the target horizontal plane as the reference point... Generate based on the reference direction One encirclement position :

[0060] (6)

[0061] in Where is the formation radius. For example, the three offset angles are:

[0062] (7)

[0063] This configuration places one interceptor directly behind the target, with two others positioned on the left and right flanks, forming a stable triangular encirclement formation. Because... As the target's position and course are updated in real time, the position of the array remains constant relative to the target's direction of movement, enabling dynamic tracking and containment.

[0064] S3. Perform optimal task allocation frame-by-frame using the Hungarian algorithm.

[0065] set up A drone, Each matrix position. Define the cost matrix. Its elements For the first Current location of the drone To the individual positions Euclidean distance:

[0066] (8)

[0067] Introducing decision variables The optimization objective is to minimize the total flight cost of all drones:

[0068] (9)

[0069] The constraint is that each drone is assigned exactly one position, and each position is assigned exactly one drone.

[0070] (10)

[0071] The Hungarian algorithm solves for the optimal match by repeatedly performing row and column reduction on the cost matrix, with a time complexity of O(n log n). The system executes once per simulation step to adapt to changes in the target position, achieving "frame-by-frame reallocation" rather than "one-time allocation," thus always ensuring the global optimal match at the current moment.

[0072] S4. Remaining Time Consistent Coordinated Guidance

[0073] Let the first The distance error from the UAV to its assigned position is:

[0074] (11)

[0075] in The array position index assigned to the Hungarian algorithm.

[0076] Define the maximum allowable approach speed as The global synchronization time is determined by the drone that is furthest away (the weakest link in the chain).

[0077] (12)

[0078] For the first A drone, in The approach speed required to eliminate position error internally is:

[0079] (13)

[0080] The approach velocity direction is a unit vector pointing towards the array position. After superimposing the target feedforward velocity, the final desired velocity command is:

[0081] (14)

[0082] in The target's current speed. If Then scale to .

[0083] This strategy ensures that all interceptors arrive at their respective positions at the same time, minimizing the target's escape space. The feedforward term enables the interceptors to follow the target's movement while approaching their positions, avoiding the systematic lag that occurs when tracking moving targets with pure position feedback.

[0084] S5. PID speed tracking control

[0085] The planning layer outputs the desired velocity vector. The underlying dynamics model takes acceleration commands as input. The PID speed controller is responsible for bridging the gap between the two.

[0086] The signal link is as follows: the planning layer outputs the desired speed, the PID speed controller calculates the speed error, and the acceleration command is synthesized through P / I / D three terms. The underlying dynamic model receives the acceleration command and applies it to the UAV through a first-order inertial link.

[0087] Speed ​​error is defined as:

[0088] (15)

[0089] Discrete calculation of three control quantities:

[0090] (16)

[0091] (17)

[0092] (18)

[0093] Final acceleration output:

[0094] (19)

[0095] in For proportional gain coefficient, For integral gain coefficient, For differential gain coefficient, For integration units, This is the saturation limit value for the integral term, intended to prevent system overshoot or instability caused by integral saturation. Parameter selection: (Powerful speed tracking) (Eliminating steady-state error) (Suppressing overshoot and oscillation). The integral term adopts... Function amplitude limiting prevents integral saturation.

[0096] S6. Artificial potential field collision avoidance

[0097] For any number frame and the first Defining a drone The position difference vector, This represents the distance between the two aircraft.

[0098] The direction of the repulsive force is a unit vector moving away from the other side.

[0099] (20)

[0100] The magnitude of the repulsive force velocity correction uses a reciprocal difference model, where the repulsive force tends to infinity as the distance gets closer.

[0101] (twenty one)

[0102] The velocity vector of the net repulsive force is the sum of the repulsive forces of each neighboring machine:

[0103] (twenty two)

[0104] The repulsive force velocity correction is added to the desired velocity to obtain the final velocity command.

[0105] (twenty three)

[0106] like Then scale proportionally to To ensure physical feasibility.

[0107] S7. Determine if the containment was successful.

[0108] A dual-determination mechanism is adopted:

[0109] Distance condition: All The interceptor aircraft simultaneously meets the requirements .

[0110] Duration condition: The cumulative duration for which the above distance conditions are met consecutively reaches [a certain value]. .

[0111] Use a counter Achieve cumulative step length:

[0112] (twenty four)

[0113] when At that time, the containment mission is considered complete.

[0114] The distance threshold for containment is typically determined based on the size of both the containment drone and the target drone. For example, if the containment drone has a diameter of 1 meter and the target drone also has a diameter of 1 meter, then... The value should be between 1.5 meters and 2 meters. This is the cumulative duration for which the distance condition is met continuously (i.e., the containment duration threshold), and is usually determined based on the actual containment needs. For example, if the target drone needs to be restricted in its movement for at least 10 minutes... The value can be set to 12 minutes.

[0115] Explanation of the differences and innovations between this invention and existing technologies

[0116] Comparison A: Proportional Guidance (PN). Difference: Proportional guidance is designed for single-machine tracking of a single target; its acceleration command is proportional to the product of the target's line-of-sight angular velocity and approach velocity. If directly applied to containment scenarios, it suffers from the following fundamental drawbacks:

[0117] Inconsistent arrival times: Each interceptor sprints independently, with closer drones arriving earlier and individually, allowing the target to break through the gap. This method ensures all interceptors arrive simultaneously through global synchronization time constraints.

[0118] Moving target tracking bias: When proportional guidance switches the tracking target to a moving containment position, the line-of-sight angle calculation object changes, resulting in systematic guidance bias. This method explicitly fuses the target feedforward velocity, fundamentally avoiding this problem.

[0119] Lack of multi-machine collaboration capability: Proportional guidance itself lacks a task allocation mechanism, and multiple machines may compete for the same position or have overlapping trajectories. This method uses the Hungarian algorithm for globally optimal assignment.

[0120] No collision avoidance mechanism: This method uses artificial potential field velocity correction for real-time collision avoidance, requiring no additional modules.

[0121] Advantages: Proportional guidance solves the problem of "a single machine chasing a single point", while this method solves the problem of "multiple machines simultaneously encircling a moving target". The two methods have fundamental differences in their problem definitions.

[0122] Comparison B: Fixed Formation Technology. Difference: The fixed formation method predefines the geometric formation, and each interceptor maintains a fixed relative position. The geometry of a fixed formation is fixed relative to world coordinates; the formation's orientation does not adjust after the target turns, resulting in a degraded encirclement effect. This method dynamically calculates the formation position based on the target's real-time heading angle, exhibiting robust adaptability to target maneuvers in any direction; fixed formations typically use a one-time static allocation, not adjusting during flight. This method recalculates the optimal allocation at each step, always maintaining the minimum global cost; facing targets with continuous spiral maneuvers, fixed formations struggle to maintain effective geometric encirclement. This method updates the formation position synchronously with the target, maintaining stable encirclement even under continuous maneuvering conditions; fixed formations lack real-time collision avoidance mechanisms during dynamic maneuvers. This method uses an artificial potential field method to correct in real-time at the velocity layer, ensuring a safe distance. Its advantages lie in: this method decouples the formation's spatial topology from the target's motion state through real-time heading perception and dynamic formation position generation, achieving true adaptive encirclement.

[0123] Comparison C: Traditional multi-machine cooperative containment methods. Difference: Traditional methods often solve problems at a single level (e.g., focusing only on guidance laws or formation control), lacking complete system integration from perception to control. This method proposes an organic combination of six technologies: RK4 high-precision dynamic integration, spiral maneuvering target modeling, Hungarian frame-by-frame optimal allocation, remaining time consistent cooperative guidance, PID velocity tracking, and artificial potential field collision avoidance, forming a three-layer closed-loop architecture of "perception-planning-control". The advantages are: the system modules have clear division of labor and distinct layers, enabling stable containment and maintenance of the encirclement posture even when the target is performing continuous spiral maneuvers.

[0124] Example

[0125] Scene design:

[0126] Case in The operation takes place in a three-dimensional space, with a total of 700 steps. The target machine starts from... Starting, it moves simultaneously in the horizontal and vertical directions with a heading angle of 45° and an elevation angle of 10°, performing a continuous spiral maneuver. The three interceptors respectively departed from take off.

[0127] Case selection parameters:

[0128] Table 1 Main parameters of the case simulation system

[0129]

[0130] Case simulation result analysis (e.g.) Figure 1 (as shown)

[0131] Driven by a spiral maneuver strategy, the target aircraft exhibits a distinctly spiraling trajectory. Under the coordinated guidance law, the three interceptor aircraft rapidly converge from their initial positions, gradually gathering towards the target aircraft's location, ultimately forming a closely aligned trajectory bundle after successfully containing it.

[0132] At the moment of successful containment, the three interceptor aircraft had completely surrounded the target aircraft, with a coordinate axis range of approximately The results were consistent with the judgment threshold, verifying the effectiveness of the dual judgment mechanism. The optimal allocation of the Hungarian algorithm ensured that the trajectories of the three machines did not intersect; the artificial potential field method guaranteed that there were no collisions among the three machines under high-density containment conditions; and the time-consistent coordination strategy ensured that the three machines completed the containment almost simultaneously, rather than approaching one another sequentially.

[0133] Deployment: First verify the stability of the initial values ​​of this method in simulation / semi-physical verification, and then gradually expand the learning.

[0134] It is understood that the present invention has been described through some embodiments, and those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.

Claims

1. A dynamic task allocation and cooperative guidance control system for the coordinated containment of mobile targets by multiple unmanned aerial vehicles (UAVs), characterized in that: include The perception layer is used to acquire the target aircraft's position, speed, and heading angle in real time. The planning layer is used to sequentially complete three tactical calculations within each control step: dynamic blocking position generation, optimal task allocation using the Hungarian algorithm, and global synchronization time coordination. The control layer converts the desired velocity into an acceleration command, superimposes it with an artificial potential field obstacle avoidance component, and then advances the physical state through a fourth-order Runge-Kutta integrator.

2. A dynamic task allocation and cooperative guidance control method for the coordinated containment of maneuvering targets by multiple unmanned aerial vehicles (UAVs), characterized in that: Includes the following steps: S1. Construct a flight dynamics model for the UAV; S2. Production containment formation: With the target's current position as the center and the target's heading angle in the horizontal plane as the reference direction, uniformly distributed containment positions are generated around it. The positions are updated in real time according to the target's position and heading. S3. Perform frame-by-frame optimal task allocation using the Hungarian algorithm, modeling task allocation as an assignment problem. Construct a cost matrix using the Euclidean distance from each machine to each array position. Call the Hungarian algorithm in real time at each step to solve for the optimal allocation, eliminate trajectory intersections, and minimize the total maneuver cost. S4. Remaining Time Consistent Cooperative Guidance: The time required for the UAV furthest from the target position is used as the global synchronization benchmark, and the remaining UAVs actively slow down and wait to ensure that all interceptors arrive at their respective positions at the same time; at the same time, the target feedforward velocity is integrated to achieve continuous tracking of moving targets. S5. PID speed tracking control, responsible for connecting the desired velocity vector and acceleration command; S6. Artificial potential field collision avoidance: The repulsive force of the artificial potential field is directly superimposed as a velocity vector correction term onto the desired guidance velocity. S7. Determine whether the containment was successful.

3. The dynamic task allocation and cooperative guidance control method for multi-UAV cooperative containment of maneuvering targets according to claim 2, characterized in that: The specific process of constructing the UAV flight dynamics model in S1 is as follows: Maintain a 9-dimensional state vector for each drone. : in These are the position, velocity, and acceleration vectors, respectively. The continuous-time dynamic equations are defined as follows: in For command acceleration, Let be the time constant of the dynamic system. Let be the derivative of position with respect to time. This is the derivative of velocity with respect to time. To prevent the drone from going out of control, an overload limit is applied to the command acceleration: in For the maximum permissible overload, It is the acceleration due to gravity. Numerical integration is performed using the fourth-order Runge-Kutta method, with the derivative function sampled four times within each step and then weighted and averaged to reduce the global truncation error to [missing value]. The four slopes of the fourth-order Runge-Kutta method are calculated as follows: The state update formula is: in For the first The update status of the step.

4. The dynamic task allocation and cooperative guidance control method for multi-UAV cooperative containment of maneuvering targets according to claim 3, characterized in that: The algorithm for generating the encirclement formation in S2 is as follows: target current position With the formation center as the reference point, and the heading angle within the target horizontal plane as the reference point... Generate based on the reference direction One encirclement position : in The radius of the formation.

5. A dynamic task allocation and cooperative guidance control method for coordinating the containment of maneuvering targets by multiple unmanned aerial vehicles (UAVs) according to claim 4, characterized in that: The specific process of S3 is as follows: set up A drone, Each position defines the cost matrix. Its elements For the first Current location of the drone To the individual positions Euclidean distance: Introducing decision variables The optimization objective is to minimize the total flight cost of all drones: The constraint is that each drone is assigned exactly one position, and each position is assigned exactly one drone. 。 6. A dynamic task allocation and cooperative guidance control method for coordinating the containment of maneuvering targets by multiple unmanned aerial vehicles (UAVs) according to claim 5, characterized in that: The strategy of S4 is as follows: Let the first The distance error from the UAV to its assigned position is: in Array position indices assigned to the Hungarian algorithm; Define the maximum allowable approach speed as The global synchronization time is determined by the drone that is furthest away: For the A drone, in The approach speed required to eliminate position error internally is: The approach velocity direction is a unit vector pointing towards the array position. After superimposing the target feedforward velocity, the final desired velocity command is: in Given the target's current speed, if Then scale to .

7. A dynamic task allocation and cooperative guidance control method for coordinating the containment of maneuvering targets by multiple unmanned aerial vehicles (UAVs) according to claim 6, characterized in that: The process of S5 is as follows: Speed ​​error is defined as: in The desired velocity vector; Discrete calculation of three control quantities: Final acceleration output: in For proportional gain coefficient, For integral gain coefficient, For differential gain coefficient, For integration units, This is the saturation limit value for the integral term, which is adopted by the integral term. Function amplitude limiting prevents integral saturation.

8. A dynamic task allocation and cooperative guidance control method for coordinating the containment of maneuvering targets by multiple unmanned aerial vehicles (UAVs) according to claim 7, characterized in that: The process of S6 is as follows: For any number frame and the first Defining a drone The position difference vector, The distance between the two aircraft. The direction of the repulsive force is a unit vector moving away from the other side. The magnitude of the repulsive force velocity correction uses a reciprocal difference model, where the repulsive force tends to infinity as the distance gets closer. The velocity vector of the net repulsive force is the sum of the repulsive forces of each neighboring machine: The repulsive force velocity correction is added to the desired velocity to obtain the final velocity command. like Then scale proportionally to .

9. A dynamic task allocation and cooperative guidance control method for coordinating the containment of maneuvering targets by multiple unmanned aerial vehicles (UAVs) according to claim 8, characterized in that: The S7 employs a dual determination mechanism: Distance condition: All The interceptor aircraft simultaneously meets the requirements , Duration condition: The cumulative duration for which the above distance conditions are met consecutively reaches [a certain value]. , Use a counter Achieve cumulative step length: when At that time, the containment mission is considered complete.

Images