A method and system for cooperative control of a UAV cluster and a UAV system

Abstract

The application discloses a UAV cluster cooperative control method and system and a UAV system, comprising: acquiring initial positions of each UAV in a UAV cluster; determining a formation member set, a first target position and an initial formation according to a received task instruction; performing path planning based on the initial positions, the first target position and the initial formation to determine a second target position corresponding to each UAV and a target formation; determining a first moving speed of the formation member set, determining a second moving speed of each UAV for maintaining the target formation based on the first moving speed, determining repulsive forces of an artificial potential field that each UAV in the formation member set is subjected to based on a relative distance, and determining a target moving speed corresponding to each UAV based on the first moving speed, the second moving speed and the repulsive forces of the artificial potential field, and controlling each UAV in the formation member set to move to the corresponding second target position according to the corresponding target moving speed. The application can improve the stability and safety of task execution.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) control, and in particular to a method, system, and UAV system for collaborative control of UAV swarms. Background Technology

[0002] With the continuous advancement of intelligent control technology, the complexity and diversity of drone missions have significantly increased. In complex environments, drone swarms can effectively adapt to environmental changes by optimizing task allocation and flight paths through collaborative mechanisms, while ensuring that drones maintain a safe distance to avoid collisions, thereby improving the stability and safety of mission execution.

[0003] Currently, UAV swarm collaborative control technology achieves swarm collaborative operations through preset behavior rules, virtual navigation, and PID control-based methods. However, these methods all have their drawbacks. Preset behavior rules, in confined spaces or environments with multiple obstacles, can easily lead to changes in the UAV swarm's topology due to obstacle avoidance, resulting in communication interruptions. Furthermore, this method has poor adaptability to large-scale swarms. Virtual navigation requires extremely high stability of the navigation point; any deviation in calculation or communication will severely impact the coordination of the entire formation. In complex environments, virtual navigation struggles to adapt to rapidly changing dynamic obstacles, compromising mission safety. PID control-based methods are highly dependent on the UAV dynamics model; when the model changes or uncertainties arise, the control effect significantly decreases, and parameter adjustment is difficult, resulting in poor adaptability in dynamic environments. Summary of the Invention

[0004] This application provides a method, system, and unmanned aerial vehicle (UAV) swarm collaborative control system to adapt to rapidly changing dynamic environments and improve the stability and safety of mission execution.

[0005] Firstly, this application provides a method for cooperative control of unmanned aerial vehicle (UAV) swarms, including:

[0006] Obtain the initial position of each drone in the drone swarm;

[0007] Based on the received mission instructions, determine the formation member set, the first target position of each drone, and the initial formation of the drone swarm.

[0008] Based on the initial positions, the first target positions, and the initial formation, path planning is performed to determine the second target position and target formation for each UAV.

[0009] A first moving speed of the formation member set is determined, a second moving speed of each UAV to maintain the target formation is determined based on the first moving speed, the artificial potential field repulsion force on each UAV in the formation member set is determined based on the relative distance, and the target moving speed corresponding to each UAV is determined based on the second moving speed and the artificial potential field repulsion force. The UAV in the formation member set is controlled to move to the corresponding second target position according to the corresponding target moving speed, wherein the relative position is the distance between each UAV and other UAVs or obstacles.

[0010] This application embodiment obtains the initial position of each drone in the drone swarm, enabling timely understanding of the drones' current positions and providing data support for subsequent dynamic adjustments. By responding to task commands, it can quickly determine the theoretical, preliminary first target position and initial formation, providing an initial formation reference for path planning and ensuring that the swarm has a clear direction at the start of the task. By determining the second target position and target formation for each drone through path planning, it can generate a second target position and target formation that better meets the actual needs of the task after considering the actual environment, ensuring that each drone in the swarm can move stably to the second target position and avoiding collisions or other safety issues caused by improper path planning. By determining the first moving speed of the swarm, it can ensure that the swarm moves as a whole towards the second target position. By detecting the relative distance between each drone and obstacles or other drones in real time, the drones can form a formation while maintaining their travel speed, obtaining a second moving speed. Based on the first moving speed, the second moving speed, and the artificial potential field repulsive force, the target moving speed can be obtained, ensuring that the drone swarm moves safely and efficiently to the second target position in complex environments, while maintaining the target formation and avoiding collisions. Compared with existing technologies, this application can adapt to rapidly changing dynamic environments and improve the stability and security of task execution.

[0011] Furthermore, the step of performing path planning based on the initial positions, the first target positions, and the initial formation to determine the second target position and target formation corresponding to each UAV specifically involves:

[0012] Calculate the first distance between the initial position of each UAV in the formation member set and the first target position, and construct a distance matrix corresponding to the formation member set based on the first distance;

[0013] The distance matrix is ​​solved with the objective of minimizing the total movement distance corresponding to the formation member set, to obtain the target allocation results between each UAV and the first target position;

[0014] Based on the target allocation result, the second target position corresponding to each UAV is determined, and the formation parameter order of each UAV in the formation member set is adjusted based on the second target position to obtain the target formation.

[0015] By determining the second target position and target formation for each UAV through path planning, a second target position and target formation that better meet the actual needs of the mission can be generated after considering the actual environment. This ensures that each UAV in the formation can move stably to the second target position, avoiding collisions or other safety issues caused by improper path planning.

[0016] Furthermore, the objective function and constraints of the distance matrix are as follows:

[0017]

[0018] In the formula, A is the total travel distance, n is the number of drones, and d ik Let x be the first distance from UAV i to the first target location k. ik Let i be a binary decision variable, where i is the index of the UAV and k is the index of the first target location.

[0019] Furthermore, determining the first moving speed of the formation member set specifically involves:

[0020] Obtain the second distance between the position vector corresponding to the virtual leader and the second target position, wherein the virtual leader is the center point of the formation member set;

[0021] Based on the second distance, a first movement speed is generated using proportional control to determine the formation member set.

[0022] By determining the first moving speed of the formation members, it can be ensured that the formation members as a whole move toward the second target position.

[0023] Furthermore, the step of determining a second movement speed for each drone to maintain the target formation based on the first movement speed specifically involves:

[0024] An adjacency matrix is ​​established based on the number of drones in the formation and preset rules;

[0025] Based on the virtual leader's position vector and the target formation, calculate the collaborative variables of each UAV;

[0026] Based on the collaborative variables and the adjacency matrix, a second moving speed for each UAV to maintain the target formation is generated through a preset consistency control algorithm.

[0027] By detecting the relative distance between each drone and obstacles or other drones in real time, the drones can form a formation while maintaining their travel speed, thus achieving a second movement speed.

[0028] Furthermore, the relevant calculation formula for the collaborative variables is as follows:

[0029] ρ i (t+1)=ρ i (t)+v i (t+1)dt;

[0030]

[0031] In the formula, ρ i (t) and ρ i (t+1) represent the position vectors of drone i at the current and next moments, respectively; N is the number of drones, with the Nth drone being the virtual leader; v i (t+1) represents the control input for the i-th UAV at the next moment; a ij This represents the communication topology between the i-th drone and the j-th drone, where 0 represents no communication topology and 1 represents a communication topology. This represents the position vector from the virtual leader to the i-th drone in the target formation, i.e., the formation parameter.

[0032] Furthermore, the formula for generating the second movement speed of each UAV to maintain the target formation through a preset consistency control algorithm is as follows:

[0033]

[0034] In the formula, u i (t) represents the control input of drone i at time t, used to directly control the second moving speed of the drone; N is the number of drones; x i (t) and x j (t) represent the position vectors of UAV i and UAV j at time t, respectively; a ij Let a be the (i,j)th item in the adjacency matrix, used to describe the communication relationship between UAV i and UAV j. If there is a communication connection between UAV i and UAV j, then a ij =1; otherwise a ij =0; i and j are both indices of the drone.

[0035] Secondly, this application provides a collaborative control system for unmanned aerial vehicle (UAV) swarms, comprising: an acquisition module, a first determination module, a second determination module, and a movement module;

[0036] The acquisition module is used to acquire the initial position of each drone in the drone cluster;

[0037] The first determining module is used to determine the formation member set, the first target position and the initial formation of each UAV in the UAV cluster according to the received task instructions;

[0038] The second determining module is used to perform path planning based on each initial position, each first target position, and the initial formation to determine the second target position and target formation corresponding to each UAV;

[0039] The movement module is configured to determine a first movement speed of the formation member set, determine a second movement speed of each UAV to maintain the target formation based on the first movement speed, determine the artificial potential field repulsion force experienced by each UAV in the formation member set based on the relative distance, and determine the target movement speed corresponding to each UAV based on the second movement speed and the artificial potential field repulsion force, and control each UAV in the formation member set to move to the corresponding second target position according to the corresponding target movement speed, wherein the relative position is the distance between each UAV and other UAVs or obstacles.

[0040] This application embodiment obtains the initial position of each drone in the drone swarm, enabling timely understanding of the drones' current positions and providing data support for subsequent dynamic adjustments. By responding to task commands, it can quickly determine the theoretical, preliminary first target position and initial formation, providing an initial formation reference for path planning and ensuring that the swarm has a clear direction at the start of the task. By determining the second target position and target formation for each drone through path planning, it can generate a second target position and target formation that better meets the actual needs of the task after considering the actual environment, ensuring that each drone in the swarm can move stably to the second target position and avoiding collisions or other safety issues caused by improper path planning. By determining the first moving speed of the swarm, it can ensure that the swarm moves as a whole towards the second target position. By detecting the relative distance between each drone and obstacles or other drones in real time, the drones can form a formation while maintaining their travel speed, obtaining a second moving speed. Based on the first moving speed, the second moving speed, and the artificial potential field repulsive force, the target moving speed can be obtained, ensuring that the drone swarm moves safely and efficiently to the second target position in complex environments, while maintaining the target formation and avoiding collisions. Compared with existing technologies, this application can adapt to rapidly changing dynamic environments and improve the stability and security of task execution.

[0041] Thirdly, this application also provides an unmanned aerial vehicle (UAV) system, including: a control console, at least one UAV, and the UAVs connected to each other via a network, wherein the UAV system can execute the UAV swarm collaborative control method as described in this application.

[0042] Furthermore, each of the aforementioned drones includes: a flight controller, an operating system, an ad hoc network module, at least one gimbal camera, and at least one lidar. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating one embodiment of the UAV swarm collaborative control method provided in this application;

[0044] Figure 2 This application provides Figure 1 A flowchart illustrating step S103 in the process;

[0045] Figure 3 This is a schematic diagram of the triangular formation of the three drones provided in this application;

[0046] Figure 4 This is a schematic diagram of the three drones in a straight line formation provided in this application;

[0047] Figure 5 This is a schematic diagram of the structure of one embodiment of the UAV swarm collaborative control system provided in this application;

[0048] Figure 6 This is a schematic diagram of one embodiment of the unmanned aerial vehicle system provided in this application. Detailed Implementation

[0049] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0050] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0051] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0052] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0053] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0054] With the development of intelligent control technology, drone missions are becoming more complex and diverse. Drone swarms optimize task allocation and paths through collaborative mechanisms, maintaining safe distances in complex environments and improving stability and safety. Existing collaborative control technologies include preset behavior rules, virtual navigation, and PID control, but each has its drawbacks. Preset behavior rules may lead to communication interruptions in narrow or obstacle-filled environments and are not suitable for large-scale swarms. Virtual navigation requires high stability of the navigation point and struggles to adapt to dynamic obstacles, impacting mission safety. PID control is highly dependent on the drone model; when the model changes or parameter adjustments are difficult, the control effect decreases, and adaptability deteriorates.

[0055] Next, the terms used in this application will be explained:

[0056] Pre-defined behavior rules are an important technical means in UAV control systems. Through predefined behavior logic, UAVs can make autonomous decisions and take actions in specific tasks.

[0057] Virtual navigation is a method for controlling drone formations. Its core idea is to introduce a virtual navigator, which can be a physically existing drone or a virtual reference point. Other drones (followers) adjust their flight attitude and position based on the virtual navigator's position, speed, and other information, thereby achieving coordinated flight of the entire formation.

[0058] PID control is a classic control algorithm widely used in the attitude and position control of unmanned aerial vehicles (UAVs). A PID controller adjusts the UAV's flight attitude through proportional, integral, and derivative components, ensuring its stability and accuracy.

[0059] The Hungarian algorithm is an algorithm for solving allocation problems, aiming to find the optimal allocation scheme that minimizes the total cost (or distance, time, etc.).

[0060] Consensus algorithms are distributed control algorithms used in multi-agent systems. They enable agents to achieve global consensus through local interactions. The key to constellation control using consensus algorithms lies in the definition of cooperative variables; each constellation configuration corresponds to a unique cooperative variable. During the design of cooperative variables, the relative positions of each UAV within the desired formation need to be determined, thus requiring a target allocation algorithm. For constellation reconfiguration problems, the key to planning and controlling constellation reconfiguration based on consensus theory lies in the switching of cooperative variables. By establishing the relationship between cooperative variables and constellation configurations—that is, each constellation configuration corresponds to a unique cooperative variable—the cooperative variables can be switched according to the corresponding configuration transformation rules, thereby enabling constellation configuration switching.

[0061] Artificial potential field (APF) is an algorithm used for robot path planning and obstacle avoidance. It guides the robot's movement by introducing a virtual potential field between the robot and the target / obstacle. In APF, the agents are treated as elastic spheres that attract each other. When two agents are far apart, they attract each other; however, when the two agents get closer, the closer their centers are, the stronger the repulsive effect, causing them to move away rapidly until they reach an equilibrium state.

[0062] Based on this, embodiments of this application provide a method, system, and unmanned aerial vehicle (UAV) swarm collaborative control system that can adapt to rapidly changing dynamic environments and improve the stability and security of mission execution.

[0063] This application provides a method, system, and unmanned aerial vehicle (UAV) swarm collaborative control method, which will be described in detail through the following embodiments. First, the UAV swarm collaborative control method in this application embodiment will be described.

[0064] The UAV swarm collaborative control method provided in this application relates to the field of UAV control. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing a UAV swarm collaborative control method, but is not limited to the above forms.

[0065] This application can also be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0066] Example 1

[0067] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the UAV swarm cooperative control method provided in this application, including steps S101 to S104.

[0068] Step S101: Obtain the initial position of each drone in the drone cluster;

[0069] In some embodiments, the current location of each drone in the drone swarm is obtained through the drone's sensors (such as GPS, IMU).

[0070] It should be noted that a drone swarm refers to a collaborative system composed of multiple drones that work together through communication and cooperation to complete complex tasks. The subsequent formation member set refers to the set of drones participating in formation flight during a specific mission; it is a subset of the drone swarm and is typically used to perform specific formation tasks. In other words, a drone swarm includes at least one or more formation member sets.

[0071] It should be noted that before obtaining the initial positions of each drone, the basic parameters of each drone need to be initialized. The specific process is as follows: receive the initialization command, initialize the basic parameters of each drone, including the drone swarm member list, formation member set, formation parameters, and initial drone positions. After initialization, wait for mission instructions.

[0072] Step S102: Based on the received task instructions, determine the formation member set, the first target position of each drone, and the initial formation in the drone swarm.

[0073] In some embodiments, a task instruction is received via a communication network. When the task instruction is received, it is parsed to identify which drones are used to perform the current task, determine how to divide the drone swarm to determine which drones in the drone swarm participate in the formation (formation member set), initially determine the target position (first target position) of each drone, and the desired formation (initial formation).

[0074] It should be noted that the information contained in the mission instructions also includes, but is not limited to, the cluster number to which the UAV belongs, the sub-cluster number, the target location, the target formation, and the spacing between formations.

[0075] It should be noted that, according to the mission instructions, the formation member list of each drone also needs to be updated to ensure that each drone knows its relationship with other drones in the current mission. Additionally, the drone swarm can be divided into several sub-clusters based on sub-cluster numbers. Each sub-cluster can execute missions independently or work collaboratively; it can form a formation or perform missions independently. This application does not impose any restrictions on this.

[0076] It should be noted that "first" and "second" do not indicate order and should be understood as nouns. The first target position is the target position assigned to each drone participating in the formation, and the subsequent second target position is the target position corresponding to each drone after path planning optimization.

[0077] It should be noted that the initial formation is the expected formation of the drone swarm initially determined in the mission instructions. It is usually a theoretical and preliminary formation design that can provide an initial formation reference for path planning (at this time, the position of each drone in the formation is not clear). The subsequent second target formation is the final formation of the drone swarm determined after path planning and optimization. It is an optimized formation (the specific position of each drone in the formation is clear), takes into account various constraints in the actual environment, and is more in line with the actual needs of mission execution.

[0078] Step S103: Based on the initial positions, the first target positions, and the initial formation, perform path planning to determine the second target position and target formation corresponding to each UAV;

[0079] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating step S103 provided in this application. Step S103 includes, but is not limited to, steps S201 to S203.

[0080] Step S201: Calculate the first distance between the initial position of each UAV in the formation member set and the first target position, and construct a distance matrix corresponding to the formation member set based on the first distance;

[0081] In some embodiments, for each UAV in the formation, the first distance between its initial position and the first target position is calculated. This can be done using the Euclidean distance formula or the Euclidean distance formula, and this application is not limited to either. Once the first distance for each UAV in the formation is calculated, all the first distances between the formation members and the first target position are organized into a distance matrix.

[0082] In some embodiments, the relevant formulas for calculating using the Euclidean distance formula are as follows:

[0083] d ik =‖diff_init[i]-(formation_list[k]+goal)‖;

[0084] In the formula, d ik This represents the distance from drone i to target position k, diff_init represents the initial position of the drone, formation_list represents the formation displacement of the drones, and goal represents the first target position.

[0085] Step S202: Solve the distance matrix with the objective of minimizing the total movement distance corresponding to the formation member set to obtain the target allocation result between each UAV and the first target position;

[0086] In some embodiments, after the distance matrix is ​​constructed, an objective function that minimizes the total movement distance needs to be established based on the distance matrix, and the constraints need to be determined. Then, the Hungarian algorithm is used to solve the problem and find the optimal target allocation scheme. By minimizing the distance between each pair of UAVs and the target position (i.e., minimizing the total movement distance), the target allocation is optimized to ensure the efficiency of task execution.

[0087] In some embodiments, the objective function and constraints of the distance matrix are specifically as follows:

[0088]

[0089] In the formula, A is the total travel distance, n is the number of drones, and d ik Let x be the first distance from UAV i to the first target location k. ik Let i be a binary decision variable, where i is the index of the UAV and k is the index of the first target location.

[0090] In some embodiments, the process of solving the objective function using the Hungarian algorithm (i.e., calculating the distance matrix as the coefficient matrix) is as follows: Step 1: Reduce the number of rows and columns of the coefficient matrix so that each row and column contains 0; ① Subtract the minimum value of each row from the minimum value of that row; ② Subtract the minimum value of each column from the minimum value of that column. Step 2: Perform trial assignment to find the optimal solution. ① After the processing in Step 1, each row and column of the coefficient matrix will contain 0 values, but it is necessary to find n independent 0s. If they are found, the optimal assignment ends; if not, continue to the following steps. ② Find the row (column) with only one 0 in the matrix and mark this 0 as... Then use the H tag Other zero values ​​in the same column (row). ③ Mark the zero values ​​in a column (row) that contain only one zero as... Then mark with H Other 0 values ​​in the same row (column). ④ Repeat steps ② and ③ until all 0s are marked. ⑤ If there are unmarked 0s in the matrix, and each row (column) has at least two 0s, select the row (column) with the fewest 0s, compare the number of 0s in the column (row) containing the 0s in that row (column), and select the column with the fewest 0s. Mark the 0s in this column, and mark other 0 values ​​in the same row and column with an H symbol. Repeat this process until all 0 elements are crossed out. ⑥ If If the number of zeros is equal to the order n of matrix C, the optimal assignment ends; otherwise, proceed to the next step. Step 3: Construct the minimum number of lines to cover all zero values, determining the maximum number of independent zeros in the matrix. ① Use × to mark zeros that are not present. 1. Mark the rows marked with × as ×. 2. Mark the columns containing 0 in the rows already marked with × as ×. 3. Mark the rows containing 0 in the columns already marked with × as ×. 4. Repeat steps 2 and 3 until no new × marks are generated. 5. Draw horizontal lines on the rows not marked with × and vertical lines on the columns marked with ×. The sum of the number of horizontal and vertical lines is the maximum number of independent 0s. Count the number of lines. If the sum of the number of lines is less than the order n of matrix C, the current coefficient matrix needs to be transformed to obtain n independent 0 elements, and proceed to the next step; if the sum of the number of lines is equal to the order n of matrix C, return to step 5 of step 2 and try again. Step 4: Reprocess the coefficient matrix. The purpose of reprocessing the coefficient matrix is ​​to increase the number of 0s. First, find the minimum value in the area not covered by lines. Then, add the minimum value to all columns marked with × and subtract the minimum value from all rows marked with ×, so as not to change the previous positions of 0s. This completes the reprocessing of the coefficient matrix. If n independent zero elements can be obtained, the optimal assignment ends; otherwise, proceed to step 2.

[0091] Step S203: Based on the target allocation result, determine the second target position corresponding to each UAV, and adjust the formation parameter order of each UAV in the formation member set based on the second target position to obtain the target formation.

[0092] In some embodiments, the second target position corresponding to each UAV is determined based on the target allocation result. Specifically, since the target allocation result has clearly defined the pairing relationship between each UAV in the formation member set and the nearest first target position, the second target position corresponding to each UAV can be directly determined based on the target allocation result.

[0093] In some embodiments, the formation parameter order of each drone in the formation member set is adjusted based on the second target position to obtain the target formation. Specifically, after determining the second target position corresponding to each drone, the formation parameter order of each drone needs to be adjusted according to the second target position. The formation parameter order refers to the arrangement order of the drones in the target formation, which determines the specific position and role of each drone in the formation. After the adjustment, the target formation can be determined. For example, assuming there are 3 drones in the formation member set, the initial formation is a triangle formation, and the initial formation parameter order is: drone 1 at the vertex, drone 2 at the lower left corner, and drone 3 at the lower right corner. If the mission instruction requires adjustment to a straight line formation, the formation parameter order may be adjusted to: drone 1 at the far left, drone 2 in the middle, and drone 3 at the far right.

[0094] It should be noted that when adjusting the order of formation parameters, it is also necessary to plan a path from the initial position to the second target position for each UAV to avoid collisions. The path planning algorithm can be, but is not limited to, A* algorithm, Dijkstra algorithm, etc. The specific path planning algorithm is not the focus of this application, so it will not be discussed here.

[0095] By adjusting the order of formation parameters, the drone swarm can quickly adapt to new mission requirements or environmental changes. At the same time, it can optimize mission allocation, path planning, and formation reconfiguration, ensuring that the drone swarm can complete missions efficiently and safely in complex environments.

[0096] By determining the second target position and target formation for each UAV through path planning, a second target position and target formation that better meet the actual needs of the mission can be generated after considering the actual environment. This ensures that each UAV in the formation can move stably to the second target position, avoiding collisions or other safety issues caused by improper path planning.

[0097] Step S104: Determine the first moving speed of the formation member set; determine the second moving speed of each UAV to maintain the target formation based on the first moving speed; determine the artificial potential field repulsion force on each UAV in the formation member set based on the relative distance; determine the target moving speed corresponding to each UAV based on the second moving speed and the artificial potential field repulsion force; control each UAV in the formation member set to move to the corresponding second target position according to the corresponding target moving speed, wherein the relative position is the distance between each UAV and other UAVs or obstacles.

[0098] In some embodiments, determining the first moving speed of the formation member set includes: obtaining a second distance between the position vector corresponding to the virtual leader and the second target position, wherein the virtual leader is the center point of the formation member set; and generating the first moving speed of the formation member set using proportional control based on the second distance. Specifically, firstly, since the position vector of the virtual leader is the center point of the formation member set, it can be obtained by calculating the average of the positions of all UAVs; secondly, after obtaining the position vector corresponding to the virtual leader, the second distance between the virtual leader's position vector and the second target position needs to be calculated using the Euclidean distance formula; then, based on the calculated second distance, the moving speed v of the virtual leader is generated using proportional control. leader And introduce the repulsive force F from the obstacle. obstacle The velocity v at the previous moment prev To update the travel speed v new The relevant formula can be: v new =v leader +F obstacle + α v prev In the formula, α is the velocity attenuation coefficient, which can be set as needed; finally, the updated velocity v... new Normalization is performed to ensure that the speed does not exceed the maximum speed limit, thereby ensuring stable movement and correct direction, so that subsequent formation members can move as a whole towards the second target position at this speed.

[0099] It should be noted that the virtual leader's movement speed is the first movement speed of the entire formation of members.

[0100] It should be noted that if the velocity is close to zero and the distance to the second target position is far, random perturbations will be introduced to avoid getting trapped in a local optimum.

[0101] By determining the first moving speed of the formation members, it can be ensured that the formation members as a whole move toward the second target position.

[0102] In some embodiments, determining the second movement speed of each drone to maintain the target formation based on the first movement speed includes: establishing an adjacency matrix according to the number of drones in the formation member set and preset rules; calculating the coordination variables of each drone based on the position vector of the virtual leader and the target formation; and generating the second movement speed of each drone to maintain the target formation based on the coordination variables and the adjacency matrix using a preset consistency control algorithm. Specifically: first, an adjacency matrix is ​​established according to the number of drones and preset rules (such as communication topology) to represent the communication relationships between drones; second, the coordination variables of each drone are calculated based on the formation parameters and the position vector of the virtual leader; then, the consistency control algorithm determines the second movement speed of each drone to maintain the target formation based on the coordination variables and the adjacency matrix.

[0103] It should be noted that the control input generated by the consistency controller will cause the cooperative variables of the UAVs in the system to tend to be consistent. When the cooperative variables of all UAVs are consistent, the target formation is formed.

[0104] In some embodiments, the relevant calculation formula for the collaborative variables is specifically as follows:

[0105] ρ i (t+1)=ρ i (t)+v i (t+1)dt;

[0106]

[0107] In the formula, ρ i (t) and ρ i (t+1) represent the position vectors of drone i at the current and next moments, respectively; N is the number of drones, with the Nth drone being the virtual leader; v i (t+1) represents the control input for the i-th UAV at the next moment; a ij This represents the communication topology between the i-th drone and the j-th drone, where 0 represents no communication topology and 1 represents a communication topology. This represents the collaborative variable pointing from the virtual leader to the i-th drone in the target formation.

[0108] It should be noted that the dynamic model of the UAV is assumed to be a first-order model, and its mathematical expression is: in, Let u represent the position vector of drone i. i (t) represents the control input of UAV i at time t. The control input u... i (t) is set to allow direct control of the speed of the controlled object; the specific control input u i (t) can be calculated using the following formula: In the formula, u i (t) represents the control input of drone i at time t, used to directly control the second moving speed of the drone; N is the number of drones; x i (t) and x j (t) represent the position vectors of UAV i and UAV j at time t, respectively; a ij Let a be the (i,j)th item in the adjacency matrix, used to describe the communication relationship between UAV i and UAV j. If there is a communication connection between UAV i and UAV j, then a ij =1; otherwise a ij =0; i and j are both indices of the UAV. If the communication network topology between multiple UAVs is an undirected graph and the undirected graph is connected, then under the action of formula (2), the first-order UAV motion model can achieve consistency. That is, when t→∞, x i →x j Valid. Substituting formula (1) into formula (2), we obtain the global controller as:

[0109]

[0110] In equation (3), D is a diagonal matrix, and L = DA is a Laplace matrix, which describes the communication topology of the UAV swarm. From equation (3), it can be seen that the closed-loop dynamic characteristics of a multi-UAV swarm depend on the Laplace matrix of the swarm. In discrete time, the method for updating the state parameters of each UAV in the swarm is as follows:

[0111] x i (t+1)=x i (t)+u i (t)*dt,

[0112] Where dt represents the sampling time interval, based on the above theoretical foundation, the formation switching problem can be transformed into a problem of coordinating variable switching, that is, a set of coordinating variables corresponds to a formation, and by designing a certain coordinating variable switching mechanism, the formation switching process can be realized.

[0113] It should be noted that the consensus control algorithm in this application employs a graph-based multi-agent consensus algorithm. By modeling agents as nodes in a graph, the algorithm uses a graph structure to represent their interactions. Through local rules and communication between agents, the interactions within the system are adjusted to achieve the consensus goal of the entire system in certain aspects. Using tools such as graph Laplacian matrices to describe the relationship graph structure, these algorithms are typically distributed. Each agent adjusts itself based on local information and communication with its neighbors to ensure the system eventually reaches a consistent state and remains stable. This approach enables multi-agent systems to self-organize and achieve consensus without central control.

[0114] By detecting the relative distance between each drone and obstacles or other drones in real time, the drones can form a formation while maintaining their travel speed, thus achieving a second movement speed.

[0115] It should be noted that the first movement speed, the second movement speed, and the target movement speed can all be understood as nouns. The first movement speed is the speed of the virtual leader, which is also the movement speed of the entire swarm, ensuring that the swarm moves towards the target point as a whole. The second movement speed is the speed generated by a consensus control algorithm based on the first movement speed, ensuring that the drone swarm moves towards the target point while maintaining the target formation. The target movement speed is the speed generated by a consensus control algorithm based on the second movement speed, ensuring that the drone swarm moves towards the target point while maintaining the target formation and avoiding collisions.

[0116] In some embodiments, the artificial potential field repulsive force experienced by each UAV in the formation member group is determined based on relative distance. Specifically, this involves: real-time detection of the distance between each UAV and other UAVs, as well as its distance from obstacles, to obtain the relative distance; then, the artificial potential field method is used to calculate the repulsive force of each UAV relative to the obstacles, i.e., the artificial potential field repulsive force. The calculation formula for the artificial potential field repulsive force is as follows:

[0117]

[0118] In the formula, The artificial potential field repulsion force between UAV i and UAV j; K is the gradient of the repulsive force field with respect to position. rep d is the gain coefficient of the artificial potential field repulsion force, used to adjust the intensity of the artificial potential field repulsion force; ij The relative distance between UAV i and UAV j; d min The minimum safe distance between drones; p i Let i be the location of the drone.

[0119] In some embodiments, the target moving speed of each UAV is determined based on the second moving speed and the artificial potential field repulsion force. Specifically, after determining the second moving speed (maintaining formation) and the artificial potential field repulsion force (considering obstacles), the two are added together to obtain the target moving speed of each UAV. That is, it can be understood that the second moving speed is adjusted in combination with the artificial potential field repulsion force to obtain the target moving speed.

[0120] In some embodiments, the formation members are controlled to move each UAV to the corresponding second target position according to the corresponding target movement speed. Specifically, the motor speed and flight attitude of each UAV are controlled according to the target movement speed, and the position and speed of the UAV are monitored in real time. Dynamic adjustments are made as needed (if an obstacle suddenly appears) to ensure that the UAV can reach the target position smoothly.

[0121] To facilitate understanding, two examples are provided to illustrate the scheme of this application:

[0122] Example 1: Assume a formation consisting of 3 drones, which sequentially complete tasks such as takeoff, target point navigation, formation adjustment, and return. First, the ground station sends an initialization command via wireless communication, and upon successful initialization, sends a confirmation signal back to the ground station. Then, the ground station instructs the drone swarm to proceed to the first target point. During flight, based on mission requirements, the ground station sends a triangular formation command (a schematic diagram of the three drones in triangular formation is shown below). Figure 3 As shown in the diagram, the drone swarm dynamically adjusts its formation according to the adjustment commands to ensure flight stability. After reaching the first target point, the ground station sends the coordinates of the second target point and adjusts the formation again to a single line (as shown in the diagram of a three-drone swarm in a single line). Figure 4 As shown, the drone swarm completes its formation change according to instructions and continues flying. After reaching the second target point, the ground station issues a return-to-base command, and the swarm begins its return journey.

[0123] Example 2: Assume a formation of 8 drones, each sequentially completing takeoff, mission execution, and return. At the start of the experiment, the ground station sends initialization commands to all drones, including attitude adjustment and sensor calibration. After initialization, the drones wirelessly transmit a confirmation signal confirming successful initialization. Next, the ground station sends the coordinates of the first target point to the drone swarm and instructs the swarm to maintain its initial formation. During flight, the ground station sends formation adjustment commands based on mission requirements, instructing the drone swarm to change to a triangular formation mid-flight. Then, the ground station issues a mission command of equal priority, containing a second target point. Based on this command, the drone swarm splits into two sub-swarms: one sub-swarm (e.g., 5 drones) heads to the first target point, and the other sub-swarm (e.g., 3 drones) heads to the second target point. Both sub-swarms execute their missions independently, reaching their respective target points along their planned paths. After completing the target point mission, the ground station sends a return command, guiding all drones to return together, ending the experiment.

[0124] This application embodiment obtains the initial position of each drone in the drone swarm, enabling timely understanding of the drones' current positions and providing data support for subsequent dynamic adjustments. By responding to task commands, it can quickly determine the theoretical, preliminary first target position and initial formation, providing an initial formation reference for path planning and ensuring that the swarm has a clear direction at the start of the task. By determining the second target position and target formation for each drone through path planning, it can generate a second target position and target formation that better meets the actual needs of the task after considering the actual environment, ensuring that each drone in the swarm can move stably to the second target position and avoiding collisions or other safety issues caused by improper path planning. By determining the first moving speed of the swarm, it can ensure that the swarm moves as a whole towards the second target position. By detecting the relative distance between each drone and obstacles or other drones in real time, the drones can form a formation while maintaining their travel speed, obtaining a second moving speed. Based on the first moving speed, the second moving speed, and the artificial potential field repulsive force, the target moving speed can be obtained, ensuring that the drone swarm moves safely and efficiently to the second target position in complex environments, while maintaining the target formation and avoiding collisions. Compared with existing technologies, this application can adapt to rapidly changing dynamic environments and improve the stability and security of task execution.

[0125] Example 2

[0126] Please refer to Figure 5 , Figure 5This is a schematic diagram of the structure of an embodiment of the UAV swarm collaborative control system provided in this application, including: an acquisition module 100, a first determination module 200, a second determination module 300, and a movement module 400;

[0127] The acquisition module 100 is used to acquire the initial position of each drone in the drone cluster;

[0128] The first determining module 200 is used to determine the formation member set, the first target position and the initial formation of each UAV in the UAV cluster according to the received task instructions;

[0129] The second determining module 300 is used to perform path planning based on each of the initial positions, each of the first target positions and the initial formation, and to determine the second target position and target formation corresponding to each UAV.

[0130] The mobile module 400 is used to determine a first moving speed of the formation member set, determine a second moving speed of each UAV to maintain the target formation based on the first moving speed, determine the artificial potential field repulsion force on each UAV in the formation member set based on the relative distance, and determine the target moving speed corresponding to each UAV based on the second moving speed and the artificial potential field repulsion force, and control each UAV in the formation member set to move to the corresponding second target position according to the corresponding target moving speed, wherein the relative position is the distance between each UAV and other UAVs or obstacles.

[0131] The information interaction and execution process between the modules in the above-mentioned UAV swarm collaborative control system are based on the same concept as the UAV swarm collaborative control method of the first aspect of the present invention, and the technical effects achieved are basically the same. For details, please refer to the description in the first embodiment of the method of the present invention, which will not be repeated here.

[0132] Example 3

[0133] Please refer to Figure 6 , Figure 6 This is a schematic diagram of an embodiment of the unmanned aerial vehicle (UAV) system provided in this application, including: a control console, at least one UAV, and the UAVs connected to each other via a network. The UAV system can execute the UAV swarm cooperative control method as described in Embodiment 1 of the present invention.

[0134] It should be noted that in the collaborative operation of a drone swarm, communication between different drones relies on ad hoc networking technology (i.e., communication between drones and between drones and the ground control station via a mesh network). Each drone can autonomously form network connections with other drones and dynamically adjust the network topology. A custom communication protocol based on UDP multicast technology has been designed to ensure that each drone in the swarm can maintain communication with other members.

[0135] It should be noted that the console is used to send task commands, monitor the status of drones, and receive data transmitted back by drones, and communicates with the drone cluster through the Mesh network; while drones exchange attitude and speed information through self-organizing network modules to achieve collaborative control and formation flight.

[0136] In some embodiments, each of the UAVs includes: a flight controller, an operating system, an ad hoc network module, at least one gimbal camera, and at least one lidar.

[0137] In some embodiments, the flight controller includes an IMU (Measurement Unit), a GPS navigation module, and an altitude hold module. The flight controller is responsible for controlling the flight state of the UAV, including attitude, speed, and position control. The IMU measures the acceleration and angular velocity of the UAV and provides attitude information. The GPS navigation module provides global positioning information for the UAV for navigation and positioning. The altitude hold module maintains the UAV at a stable flight altitude.

[0138] In some embodiments, the operating system employs Robot Operating System version 2 for communication and data exchange between internal modules of the drone. The Orange Pie 5 PLUS is the drone's main control board or computing unit, responsible for running flight control software and processing sensor data.

[0139] In some embodiments, the self-organizing network module is used to communicate with other drones and ground control consoles to build a mesh self-organizing network communication network. The gimbal camera is used to capture and transmit image data, which may be used for reconnaissance or surveillance missions. The lidar is used to measure the distance between the drone and its surroundings, providing obstacle avoidance and environmental awareness capabilities.

[0140] It should be noted that within a single drone, the various software modules within the onboard computer communicate through the ROS2 (Robot Operating System 2) system. The ROS2 system provides an efficient, stable, and flexible messaging mechanism, supporting distributed, real-time, and multi-threaded communication. Each software module publishes and subscribes to relevant messages through ROS2, thereby enabling data exchange and command transmission between the various modules within the drone.

[0141] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the method in this embodiment, depending on actual needs.

[0142] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0143] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application.

[0144] In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application for those skilled in the art.

Claims

1. A method for cooperative control of unmanned aerial vehicle (UAV) swarms, characterized in that, include: Obtain the initial position of each drone in the drone swarm; Based on the received mission instructions, determine the formation member set, the first target position of each drone, and the initial formation of the drone swarm. Based on the initial positions, the first target positions, and the initial formation, path planning is performed to determine the second target position and target formation for each UAV. A first moving speed of the formation member set is determined, a second moving speed of each UAV to maintain the target formation is determined based on the first moving speed, the artificial potential field repulsion force on each UAV in the formation member set is determined based on the relative distance, and the target moving speed corresponding to each UAV is determined based on the second moving speed and the artificial potential field repulsion force. The UAV in the formation member set is controlled to move to the corresponding second target position according to the corresponding target moving speed, wherein the relative distance is the distance between each UAV and other UAVs or obstacles. Specifically, determining the first moving speed of the formation member set involves: obtaining a second distance between the position vector corresponding to the virtual leader and the second target position, wherein the virtual leader is the center point of the formation member set; and generating and determining the first moving speed of the formation member set based on the second distance using proportional control. Specifically, determining the second movement speed of each drone to maintain the target formation based on the first movement speed involves: establishing an adjacency matrix according to the number of drones in the formation member set and preset rules; calculating the coordination variables of each drone based on the position vector of the virtual leader and the target formation; and generating the second movement speed of each drone to maintain the target formation based on the coordination variables and the adjacency matrix through a preset consistency control algorithm. The specific calculation formula for the collaborative variables is as follows: ； ； ； In the formula, and These represent the drone at the current moment and the next moment, respectively. The position vector; For the number of drones, the first A drone serves as a virtual leader; For the first The next control input for the drone; Representing the The drone and the first The communication topology of each drone, where 0 represents no communication topology and 1 represents a communication topology; Indicates the direction from the virtual leader to the first The position vector of a drone in the target formation, i.e., the formation parameter.

2. The UAV swarm cooperative control method according to claim 1, characterized in that, The path planning based on the initial positions, the first target positions, and the initial formation to determine the second target position and target formation for each UAV specifically involves: Calculate the first distance between the initial position of each UAV in the formation member set and the first target position, and construct a distance matrix corresponding to the formation member set based on the first distance; The distance matrix is ​​solved with the objective of minimizing the total movement distance corresponding to the formation member set, to obtain the target allocation results between each UAV and the first target position; Based on the target allocation result, the second target position corresponding to each UAV is determined, and the formation parameter order of each UAV in the formation member set is adjusted based on the second target position to obtain the target formation.

3. The UAV swarm cooperative control method according to claim 2, characterized in that, The objective function and constraints of the distance matrix are as follows: ； ； In the formula, It is the total distance traveled. For the number of drones, For drones To the first target location The first distance, For a binary decision variable, For drone indexing, This is the index of the first target location.

4. The UAV swarm cooperative control method according to claim 1, characterized in that, The formula for generating the second movement speed of each UAV to maintain the target formation through a preset consistency control algorithm is as follows: ； In the formula, Indicates drone In time The control input is used to directly control the second movement speed of the drone; The number of drones; and drones and drones In time The position vector; For the ()th node in the adjacency matrix , The item is used to describe the drone. and drones The communication relationship between them, if drones and drones If there is a communication connection between them, then ;otherwise ; and All of these are indexes for drones.

5. A collaborative control system for unmanned aerial vehicle (UAV) swarms, characterized in that, include: The module consists of an acquisition module, a first determination module, a second determination module, and a movement module. The acquisition module is used to acquire the initial position of each drone in the drone cluster; The first determining module is used to determine the formation member set, the first target position and the initial formation of each UAV in the UAV cluster according to the received task instructions; The second determining module is used to perform path planning based on each initial position, each first target position, and the initial formation to determine the second target position and target formation corresponding to each UAV; The movement module is configured to determine a first movement speed of the formation member set, determine a second movement speed of each UAV to maintain the target formation based on the first movement speed, determine the artificial potential field repulsion force experienced by each UAV in the formation member set based on the relative distance, and determine the target movement speed corresponding to each UAV based on the second movement speed and the artificial potential field repulsion force, and control each UAV in the formation member set to move to the corresponding second target position according to the corresponding target movement speed, wherein the relative distance is the distance between each UAV and other UAVs or obstacles; Specifically, determining the first moving speed of the formation member set involves: obtaining a second distance between the position vector corresponding to the virtual leader and the second target position, wherein the virtual leader is the center point of the formation member set; and generating and determining the first moving speed of the formation member set based on the second distance using proportional control. Specifically, determining the second movement speed of each drone to maintain the target formation based on the first movement speed involves: establishing an adjacency matrix according to the number of drones in the formation member set and preset rules; calculating the coordination variables of each drone based on the position vector of the virtual leader and the target formation; and generating the second movement speed of each drone to maintain the target formation based on the coordination variables and the adjacency matrix through a preset consistency control algorithm. The specific calculation formula for the collaborative variables is as follows: ； ； ； In the formula, and These represent the drone at the current moment and the next moment, respectively. The position vector; For the number of drones, the first A drone serves as a virtual leader; For the first The next control input for the drone; Representing the The drone and the first The communication topology of each drone, where 0 represents no communication topology and 1 represents a communication topology; Indicates the direction from the virtual leader to the first The position vector of a drone in the target formation, i.e., the formation parameter.

6. An unmanned aerial vehicle (UAV) system, characterized in that, include: The system includes a console, at least one drone, and the drones are connected via a network. The drone system can execute the drone swarm cooperative control method as described in any one of claims 1-4.

7. The unmanned aerial vehicle system according to claim 6, characterized in that, Each of the aforementioned UAVs includes: a flight controller, an operating system, an ad hoc network module, at least one gimbal camera, and at least one lidar.

Images