A multi-unmanned aerial vehicle formation control method based on packets
By classifying drones into three categories—virtual navigators, secondary navigators, and followers—and designing a damage reconstruction strategy, the problem of insufficient formation accuracy and robustness in drone formations was solved, enabling self-organized formation and formation recovery when drones are damaged.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- DALIAN UNIV
- Filing Date
- 2022-07-28
- Publication Date
- 2026-05-15
AI Technical Summary
Existing drone formation control methods suffer from low formation accuracy and insufficient system robustness when the number of drones increases, especially in the event of drone damage, they cannot effectively maintain formation.
A group-based multi-UAV formation control method is adopted, which divides UAVs into three categories: virtual navigator, secondary navigator and follower. The formation is maintained by using virtual structure and navigator-follower method, and a damage reconstruction strategy is designed to enhance the system robustness.
It improves the formation accuracy and system robustness of the formation, and can self-organize to complete identity replacement and formation transformation in the event of drone damage, making it suitable for practical applications.
Smart Images
Figure CN115167515B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of unmanned aerial vehicle (UAV) formation control technology, and more specifically to a group-based multi-UAV formation control method. Background Technology
[0002] Formation control is fundamental to maintaining orderly flight and improving the efficiency of UAV swarms. It refers to controlling the flight speed and angle of multiple UAVs flying towards a mission target to form a specific geometric shape. Furthermore, it involves constructing different formations under varying environmental conditions to achieve optimal flight efficiency. Formation control leverages the local interactions between UAVs to realize the collective behavior of the UAV swarm system, thereby achieving global formation.
[0003] Conventional formation control methods can achieve good formation results with a small number of drones, but as the number of drones increases, the lead-follow method, for example, leads to error propagation among drones, reducing formation accuracy. Furthermore, the lead-follow method requires significant communication bandwidth, and its disadvantages become more pronounced with a larger number of drones, failing to guarantee good formation results. Moreover, current formation methods do not consider the impact of the loss of a single drone on the entire formation system, a very common problem in drone applications that most current research has not addressed. Summary of the Invention
[0004] To overcome the shortcomings of low formation accuracy and low system robustness in traditional UAV swarm formation methods, this invention provides a group-based multi-UAV swarm formation control method, which can improve the formation accuracy of multi-UAV swarm formations and enhance the robustness of the swarm system.
[0005] The technical solution adopted by this invention to solve its technical problem is: a multi-UAV formation control method based on grouping, comprising the following steps:
[0006] Step 1: Establish the communication topology of the drone swarm;
[0007] Step 2: Based on the communication topology of the drone swarm, the drones are divided into three categories: virtual navigators, secondary navigators, and followers.
[0008] Step 3: Adjacent secondary navigators communicate directly and maintain flight formation using a navigator-follower method;
[0009] Step 4: A secondary navigator and several followers form a group, with the secondary navigator as the reference point, and a virtual structure method is used to maintain the flight formation.
[0010] Step 5: Based on the efficiency of the drone flight, design a formation damage reconstruction strategy.
[0011] Furthermore, let's assume there is a virtual navigator. m secondary navigators Each secondary navigator and its n followers form a group; the virtual navigator communicates directly with only one secondary navigator. The communication topology between secondary navigators is an undirected graph, while the communication topology between a secondary navigator and its followers in the group is a directed graph. Adjacent followers communicate directly with each other.
[0012] Furthermore, the secondary navigator and its n followers within the group adopt a virtual formation strategy. In the formation, the secondary navigator serves as the reference point for the formation. Based on its reference coordinates and the number of followers, the expected coordinates of each follower are generated. The followers track the expected coordinates, and then the followers maintain a fixed angle and distance from their respective secondary navigators, sailing at the same speed and yaw angle.
[0013] Furthermore, the coordinate system used by the virtual structure and the coordinate system used by the pilot-follower system both adopt the same inertial coordinate system:
[0014] The coordinates of each virtual point in the virtual structure are generated by the following formula:
[0015] (1)
[0016] in A follower yes The secondary navigator, where R is the formation distance. It's a bias term, different The values can generate different formations.
[0017] Furthermore, damage reconstruction strategies include follower damage strategies, secondary leader damage strategies, and large-scale group damage strategies.
[0018] Furthermore, the follower destruction strategy is as follows: when a follower is destroyed, the managed secondary leader records its information, and then after one step, sends the number of missing followers and their expected coordinates to the next level secondary leader with which it communicates directly. After receiving the information, the secondary leader sends the coordinates to its followers, making its followers become followers of the next level secondary leader.
[0019] Furthermore, in a communication cycle, the secondary navigator at an adjacent level only performs one follower change, and in the next communication cycle, the follower is transmitted to the upper level level level by level until it is complete; if the secondary navigator at the end has no followers, and the secondary navigator at the upper level has missing followers, the secondary navigator will become a follower and be added to the upper level.
[0020] Furthermore, the secondary leader destruction strategy is as follows: when the secondary leader is destroyed, the follower with the largest weight becomes the secondary leader of the group one communication cycle after the secondary leader is destroyed, and then updates its own weight and communication table.
[0021] Furthermore, the large-scale group destruction strategy is as follows: when there is large-scale group destruction of drones within a drone swarm, such as... When the group is destroyed, after one communication cycle, It will update its weight value Then, a new virtual navigator position is generated, and the weights of the other drones are updated in turn.
[0022] The beneficial effects of this invention include: when a drone is damaged within the drone system, the formation can still self-organize to replace the identity of the drone within the system, thereby completing the formation change, which enhances the system's robustness and makes it suitable for practical applications; within a group, the secondary navigator is used as a reference point, and a virtual structure method is used to maintain the flight formation, which improves the formation accuracy. Attached Figure Description
[0023] Figure 1 This is a secondary navigator communication topology diagram of the present invention;
[0024] Figure 2 This is a communication topology diagram of the secondary leader and follower in this invention;
[0025] Figure 3 This is a virtual structural model diagram of the present invention;
[0026] Figure 4 This is a flowchart of the damage reconstruction process of the present invention;
[0027] Figure 5 This is a reconstruction diagram of the follower damage of the present invention;
[0028] Figure 6 This is a diagram showing the damage and reconstruction of the secondary navigator in this invention.
[0029] Figure 7 This is a diagram showing the reconstruction of the damaged components of the present invention. Detailed Implementation
[0030] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0031] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0032] Example 1
[0033] Step 1. Design a communication topology for a drone swarm to achieve information consistency among all drones with minimal communication overhead:
[0034] In a swarm system, the communication topology between drones can be represented by the connectivity of a graph in graph theory. Assuming each drone is a node, and given a graph G with n nodes, .in Let be the set of vertices of the graph. Let represent the set of edges. If drone i and drone j have a communication relationship, then the edge is defined as follows: exist. Let G be the adjacency matrix of a directed graph. and If communication exists, then Otherwise, it is 0. In an undirected graph, if two elements are connected, then... .
[0035] Assume there is a virtual leader m secondary leaders Each secondary leader and its n followers Form a group. The virtual leader communicates directly with only one secondary leader. The communication structure between secondary leaders is an undirected graph, while the communication structure between the secondary leader and followers within the group is a directed graph. Adjacent followers communicate directly, such as... Figure 1 , Figure 2 As shown. If the communication topology of graph G is fixed and contains a strongly connected graph, then the nodes in graph G can asymptotically achieve consistency by controlling the input.
[0036] Step 2. Based on the communication topology of the drone swarm, the drones are divided into three categories: virtual navigators, secondary navigators, and followers.
[0037] Step 3. Adjacent secondary navigators communicate directly and use a navigator-follower method to maintain flight formation, thereby enhancing formation control.
[0038] Step 4. A secondary navigator and several followers form a group. Within the group, the secondary navigator serves as the reference point. A virtual structure method is used to maintain the flight formation and improve the formation accuracy.
[0039] In this embodiment, the secondary navigator and its n followers within the group adopt a virtual formation strategy, such as... Figure 3As shown, in the formation, the secondary navigator serves as the reference point for the formation. Based on its reference coordinates and the number of followers, it generates the desired coordinates for each follower. The followers then track these desired coordinates and maintain a fixed angle and distance from their respective secondary navigators, traveling at the same speed and yaw angle. To avoid errors caused by coordinate system transformations, the coordinate system used in the virtual structure and the navigator-follower coordinate system are both the same inertial coordinate system.
[0040] The coordinates of each virtual point in the virtual structure are generated by the following formula:
[0041] (1)
[0042] in A follower yes The secondary navigator, where R is the formation distance. It's a bias term, different The values can generate different formations.
[0043] Step 5. Finally, design a formation damage reconstruction strategy based on the efficiency of drone flight to enhance the robustness of the swarm system:
[0044] In this embodiment, there are m secondary navigators divided into m groups, and each group is managed by one secondary navigator with n followers. In complex mission scenarios with multiple drone swarms, the loss of individual drones is unavoidable. However, appropriate strategies should be set to ensure that the drone swarm maintains its formation and continues to perform the mission or reach the target point even when as many drones are lost as possible. For example... Figure 4 As shown in the damage reconstruction flowchart, this embodiment provides the following reconstruction strategies based on different damage conditions:
[0045] (1) Follower destruction and reconstruction strategy
[0046] When a follower is destroyed, the managed secondary leader records its information and then, after one step, sends the number of missing followers and their expected coordinates to the next lower-level secondary leader with which it communicates directly. Upon receiving this information, the secondary leader sends the coordinates to its followers, making them followers of the previous level's secondary leader. In one communication cycle, adjacent secondary leaders only perform one follower change, and then in the next communication cycle, followers are transmitted upwards level by level until all followers are present. If the last secondary leader has no followers, and the previous level's secondary leader has missing followers, this secondary leader will become a follower and be added to the previous level. After damage, The system will receive the information and then pass on the number of missing followers. The followers of this group will be assigned .
[0047] (2) Secondary Navigator Damage Reconstruction Strategy
[0048] When a secondary leader is destroyed, the follower with the highest weight will be promoted to secondary leader of the group after one communication cycle, and then update its own weight and communication table. For example... After damage, It will be updated to the new one. and according to Previous communication table and Establish communication.
[0049] (3) Large-scale group damage reconstruction strategy
[0050] When there is a large-scale loss of drone groups within a drone swarm, such as When the group is destroyed, after one communication cycle, It will update its weight value Then, a new virtual navigator position is generated, and the weights of the remaining drones are updated in turn.
[0051] Example 2
[0052] To verify the feasibility and effectiveness of the group-based multi-UAV formation control method, a simulation experiment was conducted on a computer with an i7-7700HQ CPU and 16GB of memory on the Matlab platform.
[0053] The drone formation changes include V-shape, vertical line, inverted V-shape, and circle. The maximum flight speed of the drones is set to 50 m / s, the maximum acceleration to 10 m / s², the interaction radius to 100 m, and the desired distance between adjacent drones to 15 m.
[0054] A formation control system for 15 drones was set up, with the drones divided into groups of 5, for a total of 3 groups. Within the time interval [0s, 10s), the drones flew in a V-shape. After 11s, one of the non-secondary leader drones in the lead group was damaged. Then, in the [10s, 15s] group, they flew in a vertical line. After 16s, the secondary leader and two other drones were damaged, and the formation flew in an inverted V-shape in the [15s, 25s] group. After 26s, the entire lead group was damaged, and the formation flew in a circular pattern after 25s. The simulation results are as follows.
[0055] from Figure 5 As can be seen, when one drone in the group was damaged, the entire drone formation was not affected. After 1 second, the group with the damaged drone re-formed, and completed the formation after 4 seconds. Figure 6As can be seen, when the secondary navigator and the other two drones in the group were destroyed, the formation changed significantly. After 2 seconds, the group with the destroyed drones began to regroup, completing the regrouping after 7 seconds. Figure 7 As can be seen, after the navigator team was destroyed, the follower team completed the role change after 2 seconds, and then completed the formation 14 seconds later, replacing the navigator team.
[0056] The simulation experiments above demonstrate that even when a drone is damaged within the drone system, the group-based drone formation control system can still self-organize to replace the drones within the system and then complete the formation change. This proves that the drone system has a certain degree of robustness and can be applied in practical applications.
[0057] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.
Claims
1. A group-based multi-UAV formation control method, characterized in that, Includes the following steps: Step 1: Establish the communication topology of the drone swarm; Step 2: Based on the communication topology of the drone swarm, the drones are divided into three categories: virtual navigators, secondary navigators, and followers. Step 3: Adjacent secondary navigators communicate directly and maintain flight formation using a navigator-follower method; Step 4: A secondary navigator and several followers form a group, with the secondary navigator as the reference point, and a virtual structure method is used to maintain the flight formation. Step 5: Based on the efficiency of the drone flight, design a formation damage reconstruction strategy; Damage reconstruction strategies include follower damage strategy, secondary leader damage strategy, and large-scale group damage strategy; The follower destruction strategy is as follows: When a follower is destroyed, the managed secondary leader records its information, and then after one step, sends the number of missing followers and their expected coordinates to the next level secondary leader with which it communicates directly. After receiving the information, the secondary leader sends the coordinates to its followers, making them followers of the previous level secondary leader. In one communication cycle, the secondary leaders of adjacent levels only perform one follower change, and in the next communication cycle, the followers are transmitted to the next level level by level until they are complete. If the last secondary leader has no followers, and the previous level secondary leader has missing followers, the last secondary leader will become a follower and be added to the previous level. The secondary leader destruction strategy is as follows: when a secondary leader is destroyed, the follower with the highest weight becomes the secondary leader of the group one communication cycle after the secondary leader is destroyed, and then updates its own weight and communication table. The large-scale group destruction strategy is as follows: when there is large-scale group destruction of drones within a drone swarm, such as... When the group is destroyed, after one communication cycle, It will update its weight value Then, a new virtual navigator position is generated, and the weights of the other drones are updated accordingly. There is 1 virtual navigator m secondary navigators Each secondary leader and its n followers form a group; The virtual navigator communicates directly with only one secondary navigator. The communication topology between secondary navigators is an undirected graph, while the communication topology between a secondary navigator and its followers within the group is a directed graph, with adjacent followers communicating directly.
2. The multi-UAV formation control method based on grouping according to claim 1, characterized in that, Within the group, the secondary navigator and its n followers adopt a virtual formation strategy. In the formation, the secondary navigator serves as the reference point for the formation. Based on its reference coordinates and the number of followers, the expected coordinates of each follower are generated. The followers track the expected coordinates, and then the followers maintain a fixed angle and distance from their respective secondary navigators, sailing at the same speed and yaw angle.
3. The multi-UAV formation control method based on grouping according to claim 2, characterized in that, The coordinate system used by the virtual structure and the coordinate system used by the pilot-follower system both adopt the same inertial coordinate system: The coordinates of each virtual point in the virtual structure are generated by the following formula: （1）； in A follower yes The secondary navigator, where R is the formation distance. It's a bias term, different The values can generate different formations.