Cluster unmanned aerial vehicle group member interaction method and device based on ad hoc network radio station, and medium

Abstract

The invention provides a cluster unmanned aerial vehicle group member interaction method and device based on an ad hoc network radio station and a medium in order to solve the problem that in the prior art, an effective task cooperation mechanism is lacked between unmanned aerial vehicles, and emergencies are difficult to deal with rapidly, and the invention provides the cluster unmanned aerial vehicle group member interaction method and device based on the ad hoc network radio station. In the dual-planning mechanism body, according to different conditions, the cluster head can reject abnormal group members, the group members can also stop the cluster head and select a new cluster head, the task cooperation capability between the unmanned aerial vehicles is enhanced, and the problem that the unmanned aerial vehicles lack an effective task cooperation mechanism and are difficult to deal with emergencies rapidly is solved. According to the method, the data synchronization of the unmanned aerial vehicle group members is carried out through the Gossip network protocol, so that the fault tolerance and reliability of data synchronization are improved, and efficient data transmission and rapid synchronization among the unmanned aerial vehicle group members are realized.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent cluster drones, and in particular relates to a method, equipment and medium for interaction between members of a cluster drone based on a self-organizing network radio station. Background Art

[0002] With the continuous advancement of unmanned system technology, drone swarms are increasingly being used, showing great potential in many fields such as disaster monitoring and logistics distribution. In order to improve work efficiency, it is necessary to build good interaction capabilities between the individual drones in the swarm (i.e., swarm members). The so-called interaction capability refers to the ability of drones to accurately exchange information, coordinate actions, avoid conflicts, and work together according to mission objectives.

[0003] In the prior art, the method for drones to interact with each other is: pre-planning + ground dynamic adjustment mode. For example, before the cluster drones take off, the cluster drones are analyzed to perform tasks and a task allocation plan is formed. During the execution of the task, the group members only interact with each other on location data, and the ground control center manually adjusts the allocation according to the changes in the task. In this way, during the execution of the task, dynamic adjustments are made through the ground end. There is a lack of effective task coordination mechanism between drones, making it difficult to respond to emergencies quickly. For example, when encountering unexpected airflow or some drones malfunction, they cannot adjust their respective flight trajectories and task execution strategies in time, which destroys the coordination of the entire cluster and reduces the success rate and work efficiency of the task. Summary of the invention

[0004] In order to solve the problem in the prior art that there is a lack of effective task coordination mechanism between drones and it is difficult to quickly respond to emergencies, the present invention provides an interaction method, device and medium between members of a cluster of drones based on a self-organizing network radio.

[0005] The technical solution adopted by the present invention is: A method for interaction between members of a cluster of drones based on a self-organizing network radio station comprises the following steps: Step S1: Pre-planning the UAV cluster mission through the ground control center, and synchronizing the pre-planned cluster mission to each member in the cluster to keep the mission information and time reference consistent; Step S2: Each group member first takes off in the route mode; then forms a cluster network through the self-organizing network radio; then, each group member synchronizes their own status information, and at the same time, selects a cluster head from the group member network to undertake the air command task of the cluster to ensure the executable capability of the cluster task; The cluster members are nodes in the cluster network; the cluster head is the central node in the cluster network; Step S3: Each group member synchronizes status information, and after the cluster head is selected, guides the remaining group members to perform cluster tasks; Step S4: When each group member performs a cluster task, the cluster head uses the ad hoc network radio to sense the group member status information and determine whether the status of the cluster member is normal; If the cluster head senses that a group member is out of contact or the status information feedback of a group member is seriously abnormal, first, the cluster head removes the abnormal group member from the existing cluster network and uses the airborne command to start task replanning. Then, each group member continues to perform the cluster task until the cluster task is fully completed. If the information of the cluster head cannot be perceived by the group members, first, the group members start the central node election procedure to select a new cluster head from the group members, and then each group member continues to perform the cluster task until the cluster task is fully completed.

[0006] Furthermore, the status information of step S2 includes: communication quality, endurance information, flight status information, and location information of the drone.

[0007] Furthermore, in step S2, when each group member synchronizes their respective status information, the protocol adopted is the Gossip network protocol.

[0008] Furthermore, in step S2, when the protocol adopted is the Gossip network protocol to synchronize the status information of each group member, different methods in the Gossip network protocol are adopted for group members who have already joined the network and group members who have just joined the network.

[0009] Furthermore, different methods in the Gossip network protocol are used for group members who have already joined the network and group members who have just joined the network, specifically: For group members who have joined the network, the Rumor-Mongering method in the Gossip network protocol is used to synchronize the status information of each new group member. For new group members, the Anti-Entropy method in the Gossip network protocol is used to synchronize the status information of each new group member.

[0010] Furthermore, in step S2, when selecting a cluster head from the group member network to undertake the air command task of the cluster, the basis for selecting the cluster head is to evaluate the weights of all group members in the group members and select the group member with the largest weight as the cluster head.

[0011] Furthermore, the expression for evaluating the weights of all group members is: ; Among them, W i is the weight of the i-th group member; ω1 is the communication quality assessment weighting coefficient; M i is the current communication quality evaluation value of the i-th group member; ω2 is the weighting coefficient of the endurance information; E i is the current endurance information evaluation value of the i-th group member; ω3 is the weighting coefficient of the flight status information; S i is the current flight status information evaluation value of the i-th group member; ω4 is the position information weighting coefficient; P i is the current location information evaluation value of the i-th group member.

[0012] Furthermore, the selection of the new cluster head in step S4 is similar to the generation of the cluster head in step S2.

[0013] Based on the same inventive concept, the present invention also provides a computer device, including: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the aforementioned method for interaction between members of a cluster of drones based on a self-organizing network radio station by executing the computer instructions.

[0014] Based on the same inventive concept, the present invention also provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the aforementioned method for interaction between members of a cluster of drones based on a self-organizing network radio.

[0015] The beneficial effects of the present invention are: In the dual planning mechanism of the present invention, according to different situations, the cluster head (central node) can eliminate abnormal group members (nodes), and the group members (nodes) can also remove the cluster head (central node) and select a new cluster head (central node), thereby enhancing the task coordination capability between drones and solving the problem of lack of effective task coordination mechanism between drones and difficulty in responding to emergencies quickly.

[0016] The present invention performs data synchronization among drone group members through the Gossip network protocol, thereby improving the fault tolerance and reliability of data synchronization and achieving efficient data transmission and rapid synchronization among drone group members. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 Schematic diagram of the process of this embodiment. DETAILED DESCRIPTION

[0019] The present invention solves the problem that there is a lack of effective task coordination mechanism between drones in the prior art and it is difficult to quickly respond to emergencies. The present invention provides a method, device and medium for interaction between members of a cluster of drones based on a self-organizing network radio. The present invention adopts a dual planning mechanism, which is reflected in that: according to different situations, the cluster head (central node) can propose abnormal group members (nodes), and group members (nodes) can also dismiss the cluster head (central node) and select a new cluster head (central node).

[0020] The disclosure below provides many different embodiments or examples to realize different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention.

[0021] The embodiments of the invention are described in detail below with reference to the accompanying drawings.

[0022] A method for interaction between members of a cluster of drones based on a self-organizing network radio station (the flowchart of this embodiment is shown in the attached figure) Figure 1 ), including the steps of: Step S1: Pre-planning the UAV cluster mission through the ground control center, and synchronizing the pre-planned cluster mission to each member in the cluster to keep the mission information and time reference consistent; Step S2: Each group member first takes off in the route mode; then forms a cluster network through the self-organizing network radio to determine whether the status of each group member in the cluster is normal; then, each group member synchronizes their own status information, and at the same time, selects a cluster head from the group member network to undertake the air command task of the cluster to ensure the executable capability of the cluster task; The cluster members are nodes in the cluster network; the cluster head is the central node in the cluster network; Step S3: Each group member (node) synchronizes status information, and after the cluster head (central node) is selected, guides the remaining group members (nodes) to perform cluster tasks; Step S4: When each group member (node) performs a cluster task, the cluster head (central node) uses the ad hoc network radio to sense the status information of the group member (node) and determines whether the status of the cluster member (node) is normal; If the cluster head (central node) senses that a group member (node) is disconnected or the status information feedback of a group member (node) is seriously abnormal, first, the cluster head (central node) removes the abnormal group member (node) from the existing cluster network and uses the airborne command to start task replanning. Then, each group member (node) continues to perform the cluster task until the cluster task is fully completed; If the information of the cluster head (central node) cannot be perceived by the group members (nodes), first, the group members (nodes) start the central node election procedure to select a new cluster head (central node) from among the group members (nodes), and then each group member (node) continues to perform the cluster task until the cluster task is fully completed.

[0023] Furthermore, the status information of step S2 includes: communication quality, endurance information, flight status information, and location information of the drone.

[0024] Furthermore, in step S2, when each group member (node) synchronizes their respective status information, the protocol adopted is the Gossip network protocol.

[0025] The reasons why the Gossip network protocol is adopted are as follows: The Gossip network protocol has high fault tolerance. In the Gossip network protocol, each node communicates with multiple other nodes, and messages are continuously transmitted between multiple nodes. The failure of individual nodes will not cause the information dissemination of the entire network to be paralyzed; the Gossip network protocol is scalable. When the number of nodes in the network increases, the Gossip protocol can still work effectively. For example, from a small network with 10 nodes to a large network with 100 nodes or even 1,000 nodes, there is no need to redesign the network structure and communication protocol on a large scale. Newly added nodes can gradually integrate into the network by communicating with existing nodes, and begin to participate in the dissemination of information and data synchronization; the Gossip network protocol can adapt to dynamic network environments. In actual applications, nodes in the network may frequently join or exit. The Gossip protocol can adapt well to this situation. When a new node joins, it can quickly obtain data from other nodes through the Anti-Entropy method to complete data synchronization.

[0026] Furthermore, in step S2, when the protocol adopted is the Gossip network protocol to synchronize the status information of each group member (node), different methods in the Gossip network protocol are adopted for the group members (nodes) that have already joined the network and the group members (nodes) that have newly joined the network.

[0027] Furthermore, different methods in the Gossip network protocol are used for group members (nodes) that have already joined the network and group members (nodes) that have newly joined the network, specifically: For group members (nodes) that have joined the network, the Rumor-Mongering method in the Gossip network protocol is used to synchronize the status information of each new group member (node); For new group members (nodes) joining the network, the Anti-Entropy method in the Gossip network protocol is used to synchronize the status information of each new group member (node) joining the network.

[0028] The reasons for using Rumor-Mongering for group members (nodes) that have joined the network are as follows: Rumor-Mongering can spread information quickly and efficiently In a dynamic environment such as a drone swarm, for the drones (nodes) that have joined the network, they have established certain communication links and network connections. The Rumor-Mongering method is like spreading rumors among a crowd. After receiving the information, each node will pass it to other adjacent nodes with a certain probability. For example, suppose there are 100 drones in the drone swarm. After each drone receives new status information (such as target location updates, environmental monitoring data changes, etc.), there is a 30% probability that the information will be passed to its neighboring drones around it. After several rounds of transmission, the information can quickly spread throughout the drone swarm, allowing all drones that have joined the network to obtain the latest status information in a timely manner.

[0029] Rumor-Mongering can reduce communication overhead Unlike the method of updating each node one by one, the Rumor-Mongering method does not require each node to communicate with all other nodes to synchronize information. It is a random, distributed communication method. When the drone swarm performs a task, such as an area search task, each drone does not need to occupy the communication channel for a long time to communicate with all other drones during the flight. In this way, while ensuring the efficiency of information dissemination, it can avoid communication congestion and resource waste caused by large amounts of data transmission, thereby reducing the overall communication overhead.

[0030] Rumor-Mongering can adapt to dynamically changing environments The position and status of drones are constantly changing during flight. Rumor-Mongering can adapt well to such dynamic changes. Because information is constantly being transmitted and updated, even if some drones temporarily lose contact or their status information is not updated in time due to environmental interference (such as electromagnetic interference, changes in weather conditions, etc.), as information continues to spread among other drones, when communication is restored, these drones can quickly obtain the latest status information, thereby ensuring that the status information of the entire drone group can be updated and synchronized in a timely manner.

[0031] The reasons why new members adopt the Anti-Entropy method are as follows: Anti-Entropy method can ensure data accuracy and consistency When a new drone joins the network, its status information may be quite different from other drones that have joined the network. The Anti-Entropy method makes the status of the new node consistent with the status of the cluster by comparing and exchanging information. Through the Anti-Entropy method, the new drone can obtain these key information from other drones, and through the data comparison and update mechanism, ensure that its status information is consistent with other drones, thereby ensuring the data accuracy and consistency of the entire drone group when performing tasks.

[0032] Anti-Entropy method for progressive data synchronization The Anti-Entropy method is a relatively gentle and gradual data synchronization method. For a newly connected drone, it may not be able to receive and process a large amount of status information at once. This method allows new drones to periodically exchange information with other drones, and only updates some data with large differences each time. For example, a new drone can first obtain the most critical flight safety rules and mission priority information, and then gradually obtain other detailed mission execution information (such as reconnaissance details requirements for a specific area, etc.) in subsequent communication cycles. This can avoid system failures or malfunctions of new drones due to information overload when joining the network.

[0033] Anti-Entropy allows new drones to be integrated into the existing network structure New drones need to be integrated into the existing drone network topology. Anti-Entropy can help new drones better understand other nodes in the network and the relationships between them. During data synchronization, new drones not only obtain status information, but also learn which drones are its main communication neighbors, which drones play a key role in the mission, and other information. This helps new drones establish their own communication links, adjust their flight attitudes and mission strategies, and better collaborate with other drones, so that they can smoothly integrate into the entire drone group and improve the collaborative work efficiency of the entire cluster.

[0034] Furthermore, in step S2, when selecting a cluster head (central node) from the group member (node) network to undertake the air command task of the cluster, the basis for selecting the cluster head (central node) is to evaluate the weights of all the group members (nodes) in the group member (node) and select the group member (node) with the largest weight as the cluster head (central node). The benefits of doing so are as follows: The member (node) with the largest weight has outstanding capabilities and leadership advantages: When the member with the largest weight is the cluster head, this member excels in certain key performance or resources. For example, it may have stronger computing power and can handle complex task allocation and flight path planning more quickly. When the cluster performs complex tasks, such as collaborative reconnaissance of a wide area, the cluster head can use its powerful computing power to quickly develop an efficient reconnaissance plan based on factors such as the location of each member, unlike weaker members who may spend too much time on task planning.

[0035] The member (node) with the largest weight as the cluster head can quickly respond to task requirements: the member with the largest weight usually also has advantages such as better communication capabilities or more energy reserves. When facing urgent task instructions, such as rapid tracking of sudden targets, good communication capabilities can ensure that it receives instructions in time and quickly conveys them to other members. Sufficient energy reserves also enable it to work stably for a long time, leading the entire cluster to complete tasks and reducing command interruptions caused by its own performance limitations.

[0036] The member (node) with the largest weight is a reliable information hub: As the node with the highest weight, the cluster head has relatively high reliability. In the process of data transmission and sharing among the cluster members, it can serve as a stable information hub. For example, it can more effectively collect data from various members and conduct preliminary screening and integration. Other members are also more inclined to trust the information transmitted by this superior cluster head, reducing cluster collaboration errors caused by information confusion or erroneous information.

[0037] Furthermore, the expression for evaluating the weights of all group members (nodes) in the group is: ; Among them, W i is the weight of the i-th group member (node); ω1 is the communication quality assessment weighting coefficient; M i is the current communication quality evaluation value of the i-th group member (node); ω2 is the weighting coefficient of the endurance information; E i is the current endurance information evaluation value of the i-th group member (node); ω3 is the weighting coefficient of the flight status information; S i is the current flight status information evaluation value of the i-th group member (node); ω4 is the position information weighting coefficient; P i is the current location information evaluation value of the i-th group member (node).

[0038] Furthermore, the selection of the new cluster head (central node) in step S4 is similar to the generation of the cluster head (central node) in step S2.

[0039] Based on the same concept, this embodiment also provides a computer device, including: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the aforementioned method for interaction between group members of a cluster of drones based on a self-organizing network radio station by executing the computer instructions.

[0040] Based on the same concept, this embodiment also provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the aforementioned method for interaction between members of a cluster of drones based on a self-organizing network radio.

Claims

1. A method for interaction between members of a swarm drone based on a self-organizing network radio, characterized in that: Includes steps: Step S1: Pre-planning the UAV cluster mission through the ground control center, and synchronizing the pre-planned cluster mission to each member in the cluster to keep the mission information and time reference consistent; Step S2: Each group member first takes off in the route mode; then forms a cluster network through the self-organizing network radio; then, each group member synchronizes their own status information, and at the same time, selects a cluster head from the group member network to undertake the air command task of the cluster to ensure the executable capability of the cluster task; The cluster members are nodes in the cluster network; the cluster head is the central node in the cluster network; Step S3: Each group member synchronizes status information, and after the cluster head is selected, guides the remaining group members to perform cluster tasks; Step S4: When each group member performs a cluster task, the cluster head uses the ad hoc network radio to sense the group member status information and determine whether the status of the cluster member is normal; If the cluster head senses that a group member is out of contact or the status information feedback of a group member is seriously abnormal, first, the cluster head removes the abnormal group member from the existing cluster network and uses the airborne command to start task replanning. Then, each group member continues to perform the cluster task until the cluster task is fully completed. If the information of the cluster head cannot be perceived by the group members, first, the group members start the central node election procedure to select a new cluster head from the group members, and then each group member continues to perform the cluster task until the cluster task is fully completed.

2. The method for interaction between members of a swarm drone based on a self-organizing network radio according to claim 1, characterized in that: The status information of step S2 includes: communication quality, endurance information, flight status information, and location information of the drone.

3. The method for interaction between members of a swarm drone based on a self-organizing network radio according to claim 1 or 2, characterized in that: In step S2, when each group member synchronizes their respective status information, the protocol adopted is the Gossip network protocol.

4. The method for interaction between members of a swarm drone based on a self-organizing network radio according to claim 3, characterized in that: In step S2, when the protocol adopted is the Gossip network protocol to synchronize the status information of each group member, different methods in the Gossip network protocol are used for group members who have already joined the network and group members who have just joined the network.

5. The method for interaction between members of a swarm drone based on a self-organizing network radio according to claim 4, characterized in that: The different methods of using the Gossip network protocol for group members who have already joined the network and group members who have just joined the network are as follows: For group members who have joined the network, the Rumor-Mongering method in the Gossip network protocol is used to synchronize the status information of each new group member. For new group members, the Anti-Entropy method in the Gossip network protocol is used to synchronize the status information of each new group member.

6. The method for interaction among members of a swarm drone based on a self-organizing network radio according to claim 1, characterized in that: In step S2, when a cluster head is selected from the group member network to undertake the air command task of the cluster, the basis for selecting the cluster head is to evaluate the weights of all group members in the group and select the group member with the largest weight as the cluster head.

7. The method for interaction among members of a swarm drone based on a self-organizing network radio according to claim 6, characterized in that: The expression for evaluating the weights of all group members is: ； Among them, W i is the weight of the i-th group member; ω1 is the communication quality assessment weighting coefficient; M i is the current communication quality evaluation value of the i-th group member; ω2 is the weighting coefficient of the endurance information; E i is the current endurance information evaluation value of the i-th group member; ω3 is the weighting coefficient of the flight status information; S i is the current flight status information evaluation value of the i-th group member; ω4 is the position information weighting coefficient; P i is the current location information evaluation value of the i-th group member.

8. The method for interaction between members of a swarm drone based on a self-organizing network radio according to any one of claims 4 to 7, characterized in that: The method of selecting the new cluster head in step S4 is similar to the method of generating the cluster head in step S2.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to execute a method for interaction between members of a cluster of drones based on a self-organizing network radio as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute a method for interaction between members of a cluster of drones based on a self-organizing network radio as described in any one of claims 1 to 8.

Images