A method and system for controlling drone swarm formations

Abstract

The present invention discloses a method and system for controlling a formation of a drone cluster, and relates to the technical field of drones. The method comprises obtaining cluster information of a drone cluster; determining a formation keeping control amount and an obstacle rejection control amount to which the drone is subjected according to the cluster information, and determining a comprehensive control amount according to the formation keeping control amount and the obstacle rejection control amount; inputting the comprehensive control amount into a control law model to obtain a control input, and controlling the drone according to the control input. The present invention fully considers the relevant information of the drone and the obstacle, designs two control amounts, namely, a formation keeping control amount and an obstacle rejection control amount, and fully considers the needs of formation and obstacle avoidance. At the same time, it also considers the interaction between drones and the influence of environmental factors, and improves the computing efficiency and model accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a method and system for controlling a cluster formation of unmanned aerial vehicles. Background Art

[0002] With the rapid development of Internet technology and artificial intelligence technology, distributed sensor networks are widely used to connect independent subsystems into efficient large systems. This trend is particularly evident in the field of military technology. Through the collaborative work of multiple drones, drone formations have significantly improved their ability to perceive the surrounding environment, their ability to resist interference, and the robustness of mission completion. Especially in terms of multi-target detection and identification, and saturation attacks under high-intensity and high-precision confrontation, drone formations have demonstrated advantages that cannot be matched by a single drone, and have become a force that cannot be ignored in new-quality warfare.

[0003] However, UAV formations face many challenges in the collaborative mission process, such as a large number of nodes, strong dynamic changes in the process, strong coupling between nodes, high uncertainty, and many task constraints. These characteristics put forward huge requirements for the control of the formation of UAV clusters, especially under the constraints of mass, space, power consumption, cost, etc. The storage space and computing power of a single-node UAV are limited. How to achieve effective control of cluster UAV formations in a strong dynamic, strong uncertainty, and strong coupling environment with limited computing power has become an urgent problem to be solved in the current UAV formation control.

[0004] At present, the commonly used methods for formation control of drone clusters include leader-follow methods, and formation control methods designed based on principles such as behavior, virtual structure, graph theory, and artificial potential field. These methods have their own advantages and disadvantages and are applicable to different types of formation control problems. For example, patent CN113238583B proposes a method for dense formation flight and collision avoidance control of fixed-wing drones. The relative navigation and positioning information of each drone in the drone formation is obtained through visual positioning, and then the formation configuration maintenance control quantity and the collision avoidance control quantity between drones are constructed, and a fixed-focus monocular vision camera is used to measure obstacles and avoid them, and the obstacle avoidance control quantity is obtained. The method also determines the overall path tracking control quantity of the formation based on the relative navigation and positioning information, and finally realizes the dense formation flight and collision avoidance control of the drone formation by fusing these control quantities.

[0005] Although the prior art CN202110792897.6 proposes a control method based on visual positioning and a monocular visual camera in the dense formation flight and collision avoidance control of fixed-wing UAVs, and realizes formation configuration maintenance, collision avoidance between aircrafts and obstacle avoidance control, this method still has the following disadvantages:

[0006] 1) High computing power requirements

[0007] First, existing technologies require real-time processing of large amounts of image data for accurate visual positioning and obstacle avoidance. This requires drones to be equipped with high-performance processors and large-capacity storage space to support complex algorithm operations and data storage. However, on drones with limited computing power, it may not be possible to process this data in real time, resulting in inaccurate positioning or slow obstacle avoidance response.

[0008] For example, when flying at high speed or in complex environments, drones need to quickly capture and process large amounts of image information to make timely obstacle avoidance decisions. If computing power is insufficient, the drone may not be able to identify and avoid obstacles in a timely manner, increasing the risk of collision.

[0009] Secondly, the prior art is to find the drone closest to the obstacle based on the position information of the obstacle relative to the drone and use the drone closest to the obstacle for detection, and define the second artificial potential function with the distance between the detected drone closest to the obstacle and the obstacle as a parameter, and apply a repulsive control effect to the drone closest to the obstacle based on the second artificial potential function to implement obstacle avoidance control. The repulsive control effect applied to implement obstacle avoidance control is the obstacle avoidance control amount. If several obstacles are at the same distance from the same drone or the same obstacle is at the same distance from the same drone, the algorithm needs to apply multiple repulsive control effects, resulting in low control efficiency.

[0010] 2) Poor control accuracy

[0011] Existing methods may not fully consider factors such as the impact of airflow between drones, the dynamics of obstacles, and non-drone disturbances (such as weather disturbances). This results in inaccurate control models that cannot accurately predict and respond to complex situations in actual flight.

[0012] For example, in formation flying, airflow interference between drones may affect each other's flight stability and control accuracy. If the control model does not fully consider this factor, it may lead to an increase in the relative position deviation between drones, or even a collision. Similarly, if the dynamics of obstacles (such as trajectory prediction of moving vehicles) and non-aircraft disturbances (such as sudden weather changes) are not fully considered, drones may not be able to avoid obstacles or adapt to environmental changes in time, thereby increasing flight risks. Summary of the invention

[0013] In view of the defects in the prior art, the present invention provides a method and system for controlling a swarm of unmanned aerial vehicles (UAVs), which solves the defects in the prior art.

[0014] In one aspect, the present invention provides a method for controlling a drone cluster formation, the method comprising:

[0015] Acquire cluster information of the drone cluster, wherein the cluster information includes the number of drones in the drone cluster and initial information of each drone, wherein the initial information includes a drone position vector, a velocity vector, a mass, and a position vector of an obstacle;

[0016] Determine a formation keeping control amount and an obstacle rejection control amount for the UAV according to the cluster information, and determine a comprehensive control amount according to the formation keeping control amount and the obstacle rejection control amount;

[0017] The comprehensive control quantity is input into the control law model to obtain a control input, and the UAV is controlled according to the control input.

[0018] Preferably, determining the amount of formation holding control of the UAV according to the cluster information includes:

[0019] Determine the drone based on the position vector and velocity vector With drones The Euclidean distance and velocity difference between them;

[0020] The formation keeping control amount of the target UAV is determined according to the Euclidean distance and the speed difference.

[0021] Preferably, the formula for determining the amount of formation holding control that the UAV is subjected to according to the cluster information is:

[0022]

[0023] In the formula, Indicates the amount of control the formation maintains. represents the neighbor set of drone i, represents the position vector of drone i, represents the velocity vector of drone i, represents the position vector of UAV j, represents the velocity vector of UAV j, Represents the weight coefficient related to the relative position and speed between UAV i and UAV j, represents the desired relative position vector, , Both represent parameters for adjusting the compactness of the formation. represents the Euclidean distance between UAV i and UAV j, represents the speed difference between UAV i and UAV j.

[0024] Preferably, determining the obstacle rejection control amount of the drone according to the cluster information includes:

[0025] Determine the distance between the UAV and the obstacle based on the UAV position vector and the obstacle position vector, and determine the prediction time step and obstacle weight coefficient based on the distance between the UAV and the obstacle and the UAV speed vector;

[0026] Determining an expected distance between the UAV and the obstacle after the prediction time step according to the position vector, the velocity vector of the UAV, the position vector of the obstacle and the prediction time step;

[0027] An obstacle rejection control amount is determined according to the expected distance and the obstacle weight coefficient.

[0028] Preferably, the formula for determining the obstacle rejection control amount of the drone according to the cluster information is:

[0029]

[0030] In the formula, represents the obstacle rejection control amount, represents the position vector of obstacle k, represents the prediction time step, Represents the parameter for adjusting the obstacle avoidance sensitivity. represents the obstacle weight coefficient, Indicates that UAV i is in the pre-time step The expected distance between the rear and obstacle k.

[0031] Preferably, the formula for determining the comprehensive control amount according to the formation keeping control amount and the obstacle rejection control amount is:

[0032]

[0033] In the formula, represents the comprehensive control quantity, is a constant.

[0034] Preferably, the comprehensive control quantity is input into a control law model to obtain a control input, and before controlling the UAV according to the control input, the method includes: considering the influence of airflow between UAVs and constructing a control input law model.

[0035] Preferably, the control input law model is:

[0036]

[0037] In the formula, represents the control input of drone i, represents the expected acceleration of UAV i, represents non-inter-machine disturbance, It represents the inter-machine disturbance between UAV i and UAV s. The relative distance between UAV s and UAV i is less than the set value. The non-inter-machine disturbance includes weather disturbance.

[0038] Preferably, the formula for calculating the inter-machine interference is:

[0039] .

[0040] On the other hand, a UAV swarm formation control system includes:

[0041] An acquisition module is used to acquire cluster information of a drone cluster, wherein the cluster information includes the number of drones in the drone cluster and initial information of each drone, wherein the initial information includes a drone position vector, a velocity vector, a mass, and a position vector of an obstacle;

[0042] A calculation module is used to determine the formation keeping control amount and obstacle rejection control amount of the UAV according to the cluster information, and determine the comprehensive control amount according to the formation keeping control amount and obstacle rejection control amount.

[0043] The control module is used to input the comprehensive control quantity into the control law model to obtain the control input, and control the UAV according to the control input.

[0044] The beneficial effects of the present invention are embodied in: an embodiment of the present invention provides a method for controlling a formation of a drone cluster, the method comprising obtaining cluster information of a drone cluster, the cluster information comprising the number of drones in the drone cluster and the initial information of each drone, the initial information comprising the drone position vector, velocity vector, mass and position vector of the obstacle; determining the formation keeping control amount and obstacle rejection control amount of the drone according to the cluster information, determining the comprehensive control amount according to the formation keeping control amount and obstacle rejection control amount; inputting the comprehensive control amount into a control law model to obtain a control input, and controlling the drone according to the control input. The present invention fully considers the relevant information of the drone and the obstacle, designs two control amounts, namely the formation keeping control amount and the obstacle rejection control amount, fully considers the formation needs and the obstacle avoidance needs, and at the same time, also considers the interaction between drones and the influence of environmental factors, improves the computing efficiency and model accuracy, and increases the flexibility of control.

[0045] In addition, the present invention also provides a UAV cluster formation control system, which includes an acquisition module, a calculation module and a control module, thereby realizing effective control of the UAV cluster formation. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0047] Figure 1 A flowchart of a method for controlling a drone cluster formation provided by an embodiment of the present invention;

[0048] Figure 2 A flow chart of a method for determining a formation holding control amount provided by an embodiment of the present invention;

[0049] Figure 3 A flowchart of a method for determining an obstacle avoidance and rejection control amount provided by an embodiment of the present invention;

[0050] Figure 4 A system block diagram of a drone swarm formation control system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0052] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0053] Example 1

[0054] like Figure 1 As shown, the present invention provides a method for controlling a drone cluster formation, the method comprising:

[0055] Step 1: Obtain cluster information of the drone cluster, wherein the cluster information includes the number of drones in the drone cluster and initial information of each drone, wherein the initial information includes a drone position vector, a velocity vector, a mass, and a position vector of an obstacle.

[0056] First, it is necessary to obtain detailed information about the drone cluster, including but not limited to the number of drones in the cluster, the initial information of each drone (specifically including the position vector, velocity vector, mass of the drone), and the position vector of the obstacle. The accuracy of this information is directly related to the formulation and execution of subsequent control strategies.

[0057] Specifically, the position vector and velocity vector of the drone can reflect the real-time state and dynamic changes of the drone in space; the mass information helps the system to more accurately predict the motion law and response characteristics of the drone; and the position vector of the obstacle is an important basis for formulating obstacle avoidance strategies. By comprehensively and accurately obtaining this information, the system can lay a solid foundation for subsequent formation maintenance and obstacle rejection control.

[0058] Through this step, complete information about the drone cluster can be obtained, providing strong data support for the subsequent formulation of control strategies.

[0059] Step 2: determining the formation keeping control amount and obstacle rejection control amount of the UAV according to the cluster information, and determining the comprehensive control amount according to the formation keeping control amount and obstacle rejection control amount.

[0060] like Figure 2 As shown, in an embodiment of the present invention, determining the amount of formation holding control of the drone according to the cluster information includes: determining the amount of formation holding control of the drone according to the position vector and the velocity vector With drones The Euclidean distance and speed difference between the two targets; and determining the formation keeping control amount of the target UAV according to the Euclidean distance and speed difference.

[0061] In an embodiment of the present invention, the formula for determining the amount of formation holding control that the drone is subjected to according to the cluster information is:

[0062]

[0063] In the formula, Indicates the amount of control the formation maintains. represents the neighbor set of drone i, represents the position vector of drone i, represents the velocity vector of drone i, represents the position vector of UAV j, represents the velocity vector of UAV j, Represents the weight coefficient related to the relative position and speed between UAV i and UAV j, represents the desired relative position vector, , Both represent parameters for adjusting the compactness of the formation. represents the Euclidean distance between UAV i and UAV j, represents the speed difference between UAV i and UAV j.

[0064] The calculation of the formation-keeping control amount takes into account the relative position and speed between the UAVs. By calculating the Euclidean distance and speed difference between the UAVs, the expected relative position and speed that the UAVs should maintain can be obtained. This calculation method can ensure that the UAVs always maintain the predetermined formation shape during flight, and maintain the stability and consistency of the formation even in a dynamically changing environment.

[0065] like Figure 3 As shown, in an embodiment of the present invention, the obstacle rejection control amount of the UAV is determined according to the cluster information, including: determining the distance between the UAV and the obstacle according to the UAV position vector and the obstacle position vector, and determining the prediction time step and the obstacle weight coefficient according to the distance between the UAV and the obstacle and the UAV speed vector; determining the expected distance between the UAV and the obstacle after the prediction time step according to the UAV position vector, speed vector, obstacle position vector and the prediction time step; determining the obstacle rejection control amount according to the expected distance and the obstacle weight coefficient.

[0066] The prediction time step is usually determined based on the flight speed of the drone, the speed of the obstacle, and the real-time requirements of the obstacle avoidance system. A shorter prediction time step can provide more accurate trajectory prediction results, but may increase the computational complexity; a longer prediction time step may reduce the computational complexity, but may sacrifice a certain degree of prediction accuracy. Therefore, it is necessary to find a balance between prediction accuracy and computational complexity.

[0067] In an embodiment of the present invention, the formula for determining the obstacle rejection control amount of the drone according to the cluster information is:

[0068]

[0069] In the formula, represents the obstacle rejection control amount, represents the position vector of obstacle k, represents the prediction time step, Represents the parameter for adjusting the obstacle avoidance sensitivity. represents the obstacle weight coefficient, Indicates that UAV i is in the pre-time step The expected distance between the rear and obstacle k.

[0070] The calculation of obstacle rejection control takes into account the distance and speed between the drone and the obstacle. By predicting the possible collision between the drone and the obstacle in the future, the corresponding rejection control can be calculated in advance, allowing the drone to adjust the flight trajectory in advance to avoid collision. This ability to predict and avoid collisions greatly improves the flight safety of drone clusters.

[0071] It should be noted that the embodiment of the present invention regards the obstacle as a static obstacle. In some preferred embodiments, the dynamics of the obstacle can also be considered. In this case, the prediction time step and the obstacle weight coefficient are determined according to the distance between the UAV and the obstacle, the UAV speed vector and the obstacle speed vector. And the position vector of the obstacle in the calculation formula of the obstacle rejection control amount is modified to the position vector after the preset time step.

[0072] In the embodiment of the present invention, the formula for determining the comprehensive control amount according to the formation keeping control amount and the obstacle rejection control amount is:

[0073]

[0074] In the formula, represents the comprehensive control quantity, is a constant.

[0075] It should be noted that the embodiment of the present invention also introduces weight coefficients and adjustment parameters in the calculation process to more flexibly respond to different flight environments and mission requirements. For example, by adjusting the parameters of the formation tightness, the relative distance between drones can be changed according to the mission requirements, thereby achieving a tighter formation or a looser formation. Similarly, by adjusting the parameters of the obstacle avoidance sensitivity, the response speed of the drone to obstacles can be changed to adapt to different flight speeds and obstacle densities.

[0076] Compared with the prior art, the present invention fully considers the interaction between drones and the influence of environmental factors when calculating the formation maintenance control amount and the obstacle rejection control amount. By dynamically adjusting the control amount, more precise control of drone formation flight can be achieved, improving the stability and obstacle avoidance capability of the formation.

[0077] Step 3: Input the comprehensive control quantity into the control law model to obtain the control input, and control the UAV according to the control input.

[0078] Specifically, the control input is a vector of thrust or control force, which determines the acceleration change of the drone. The drone adjusts its flight state according to the control input to achieve the goals of formation keeping and obstacle avoidance. After receiving the control input, the drone's control system converts it into specific control instructions such as motor speed and propeller thrust. These control instructions are sent to the drone's actuators, such as motors and propellers, to adjust the drone's flight state.

[0079] In an embodiment of the present invention, the comprehensive control quantity is input into a control law model to obtain a control input, and before controlling the UAV according to the control input, the method includes: considering the influence of airflow between UAVs and constructing a control input law model.

[0080] In the embodiment of the present invention, the control input law model is:

[0081]

[0082] In the formula, represents the control input of drone i, represents the expected acceleration of UAV i, represents non-inter-machine disturbance, It represents the inter-machine disturbance between UAV i and UAV s. The relative distance between UAV s and UAV i is less than the set value. The non-inter-machine disturbance includes weather disturbance.

[0083] In the embodiment of the present invention, the formula for calculating the inter-machine interference is:

[0084] .

[0085] It should be noted that the inter-machine interference is a simple calculation model provided by an embodiment of the present invention. In some preferred embodiments, the interval between the UAVs and the disturbing UAVs that cause disturbances to the UAVs can also be considered to correct the inter-machine disturbance. As the distance between the UAV and the corresponding disturbing UAV becomes longer, the inter-machine disturbance attenuation increases, and the greater the number of interval UAVs, the greater the disturbance attenuation rate.

[0086] After determining the formation keeping control amount and obstacle rejection control amount, these two control amounts need to be combined to form the final comprehensive control amount. The calculation of the comprehensive control amount takes into account the needs of both formation keeping and obstacle avoidance. By setting a reasonable weight coefficient, a balance between the needs of these two aspects can be achieved.

[0087] When constructing the control law model, the influence of airflow between drones is fully considered. Since drones will generate airflow disturbances during flight, this disturbance will affect the flight status of other drones. Therefore, when constructing the model, the concept of inter-drone disturbances is introduced, and corresponding algorithms are used to calculate and compensate for this disturbance. At the same time, the influence of non-inter-drone disturbances (such as weather disturbances) is also considered to ensure that the model can more accurately reflect the actual flight status of the drone.

[0088] Compared with the prior art, the present invention fully considers the complexity and dynamics of UAV formation flight when calculating the comprehensive control quantity and control input. By comprehensively considering the needs of formation maintenance, obstacle avoidance, and interaction between UAVs, more precise control of UAV formation flight can be achieved. This can not only improve the stability and flight efficiency of the formation, but also ensure the safe flight of UAVs in complex environments.

[0089] In summary, an embodiment of the present invention provides a method for controlling a formation of a drone cluster, the method comprising obtaining cluster information of a drone cluster, the cluster information comprising the number of drones in the drone cluster and the initial information of each drone, the initial information comprising the drone position vector, velocity vector, mass and position vector of the obstacle; determining the formation keeping control amount and obstacle rejection control amount of the drone according to the cluster information, determining the comprehensive control amount according to the formation keeping control amount and obstacle rejection control amount; inputting the comprehensive control amount into a control law model to obtain a control input, and controlling the drone according to the control input. The present invention fully considers the relevant information of the drone and the obstacle, designs two control amounts, namely the formation keeping control amount and the obstacle rejection control amount, fully considers the formation needs and the obstacle avoidance needs, and at the same time, also considers the interaction between drones and the influence of environmental factors, improves the computing efficiency, dynamic adaptability and model accuracy, and increases the flexibility of control.

[0090] Example 2

[0091] like Figure 4 As shown, an embodiment of the present invention provides a drone cluster formation control system, the system comprising:

[0092] The acquisition module is used to obtain cluster information of the drone cluster, wherein the cluster information includes the number of drones in the drone cluster and the initial information of each drone, wherein the initial information includes the drone position vector, velocity vector, mass and obstacle position vector.

[0093] The calculation module is used to determine the formation keeping control amount and obstacle rejection control amount of the UAV according to the cluster information, and determine the comprehensive control amount according to the formation keeping control amount and obstacle rejection control amount.

[0094] The control module is used to input the comprehensive control quantity into the control law model to obtain the control input, and control the UAV according to the control input.

[0095] Specific working process: The acquisition module is started first, and the cluster information of the drone cluster is collected in real time through the integrated sensor system (such as GPS, radar, camera, etc.) and communication equipment. This information includes the total number of drones in the drone cluster and the initial information of each drone. The initial information describes in detail the position vector (i.e., its coordinates in three-dimensional space), velocity vector (i.e., the direction and speed of its movement), mass (affecting the dynamic characteristics of the drone), and the position vector of the surrounding obstacles (used to avoid collision) of each drone. These data are the basis for subsequent calculations and controls, ensuring the accuracy and real-time performance of the system. After receiving the cluster information from the acquisition module, the calculation module begins to process these data. First, according to the position vector and velocity vector of the drone, as well as the target shape and speed of the formation, the formation control amount that each drone needs to maintain is calculated. At the same time, according to the position vector of the obstacle and the current position of the drone, the obstacle rejection control amount is calculated to avoid collision between the drone and the obstacle. Finally, the formation maintenance control amount and the obstacle rejection control amount are comprehensively considered to determine the comprehensive control amount of each drone. Finally, after the control module receives the comprehensive control amount from the calculation module, it inputs it into the pre-designed control law model. The control law model calculates specific control inputs based on the comprehensive control quantity, such as the speed of the motor, the angle of the servo, etc. These control inputs are sent to the actuators of the drone, such as the motor and servo, to control the flight of the drone.

[0096] It should be understood that the drone cluster formation control system provided in the embodiment of the present invention and the drone cluster formation control method provided in the above embodiment are based on the same inventive concept. For more specific working principles of each module in the embodiment of the present invention, please refer to the above embodiment and will not be repeated in the embodiment of the present invention.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for controlling a drone swarm formation, characterized in that: include: Acquire cluster information of the drone cluster, wherein the cluster information includes the number of drones in the drone cluster and initial information of each drone, wherein the initial information includes a drone position vector, a velocity vector, a mass, and a position vector of an obstacle; Determine a formation keeping control amount and an obstacle rejection control amount for the UAV according to the cluster information, and determine a comprehensive control amount according to the formation keeping control amount and the obstacle rejection control amount; Inputting the comprehensive control quantity into a control law model to obtain a control input, and controlling the UAV according to the control input; Determining the amount of formation holding control that the UAV is subjected to according to the cluster information includes: Determine the Euclidean distance and speed difference between UAV i and UAV j according to the position vector and the speed vector; determining a formation keeping control amount of the target UAV according to the Euclidean distance and the speed difference; The formula for determining the amount of formation-keeping control that the UAV is subjected to based on the cluster information is: In the formula, V1(p i ,v i ) represents the formation maintenance control quantity, N(i) represents the neighbor set of UAV i, and p i represents the position vector of drone i, v i represents the velocity vector of drone i, p j represents the position vector of UAV j, v j represents the velocity vector of UAV j, A ij represents the weight coefficient related to the relative position and speed between UAV i and UAV j, r dij represents the desired relative position vector, σ p , σ v Both represent parameters for adjusting the compactness of the formation, ||p i -p j || represents the Euclidean distance between UAV i and UAV j, ||v i -v j || represents the speed difference between drone i and drone j.

2. A method for controlling a drone swarm formation according to claim 1, characterized in that: Determining the obstacle rejection control amount of the UAV according to the cluster information includes: Determine the distance between the UAV and the obstacle based on the UAV position vector and the obstacle position vector, and determine the prediction time step and obstacle weight coefficient based on the distance between the UAV and the obstacle and the UAV speed vector; Determining an expected distance between the UAV and the obstacle after the prediction time step according to the position vector, the velocity vector of the UAV, the position vector of the obstacle and the prediction time step; An obstacle rejection control amount is determined according to the expected distance and the obstacle weight coefficient.

3. A method for controlling a drone swarm formation according to claim 2, characterized in that: The formula for determining the obstacle rejection control amount of the UAV according to the cluster information is: In the formula, V2(p i ,v i ) represents the obstacle rejection control amount, o k represents the position vector of obstacle k, Δt k represents the prediction time step, σ o Represents the parameter for adjusting the obstacle avoidance sensitivity, C k represents the obstacle weight coefficient, ||p i -o k -v i Δt k || indicates the distance of UAV i in the pre-time step Δt k The expected distance between the rear and obstacle k.

4. A method for controlling a drone swarm formation according to claim 2, characterized in that: The formula for determining the comprehensive control amount based on the formation keeping control amount and the obstacle rejection control amount is: In(p i, in i )=aV1(p i, in i )+(1-a)V2(p i, in i ) In the formula, V(p i ,v i ) represents the comprehensive control quantity, and a is a constant.

5. The method for controlling a drone swarm formation according to claim 1, characterized in that: The comprehensive control quantity is input into the control law model to obtain the control input. Before controlling the UAV according to the control input, the method includes: considering the influence of airflow between the UAVs and constructing a control input law model.

6. A method for controlling a drone swarm formation according to claim 5, characterized in that: The control input law model is: In the formula, u i represents the control input of drone i, a di represents the expected acceleration of UAV i, f i represents non-inter-machine disturbance, f is It represents the inter-machine disturbance between UAV i and UAV s. The relative distance between UAV s and UAV i is less than the set value. The non-inter-machine disturbance includes weather disturbance.

7. A method for controlling a drone swarm formation according to claim 6, characterized in that: The formula for calculating the inter-machine interference is:

8. A UAV swarm formation control system, characterized in that: The control method according to claim 1 comprises: An acquisition module is used to acquire cluster information of a drone cluster, wherein the cluster information includes the number of drones in the drone cluster and initial information of each drone, wherein the initial information includes a drone position vector, a velocity vector, a mass, and a position vector of an obstacle; A calculation module is used to determine the formation keeping control amount and obstacle rejection control amount of the UAV according to the cluster information, and determine the comprehensive control amount according to the formation keeping control amount and obstacle rejection control amount. The control module is used to input the comprehensive control quantity into the control law model to obtain the control input, and control the UAV according to the control input.

Images