A converged distribution resonant communication method for multi-unmanned system collaboration

Abstract

A convergence and distribution resonant communication method for the coordination of multiple unmanned systems, comprising the following steps: a relay drone uses a positioning system to calculate the distance from its own position to a convergence point S; the relay drone determines which regional node it belongs to based on the distance from itself to the convergence point S, and determines which sub-period beat it currently belongs to; each level of cellular nodes completes the data transmission process in one cycle; determines whether the node is accessible; if yes, executes the next step; if no, returns to recalculate the distance from the relay drone to the convergence point S; sends data until the packet is sent or the safe distance is reached, returns to the original position, and ends. The present invention can form a dynamic resonant communication link in succession according to predetermined procedures and rules without the need for theoretical calculation of point layout and manual intervention, so that multiple unmanned systems cooperate with each other to complete the periodic convergence and / or distribution of information.

Description

Technical Field

[0001] The present invention relates to the technical field of relay communication of unmanned aerial vehicle systems, and in particular to the technical field of a convergence distribution resonant communication method for coordination of multiple unmanned aerial vehicle systems. Background Art

[0002] The effective aggregation and distribution of data is one of the eternal themes of communication networks, especially for sensor networks or broadcast announcement information distribution in networks. In such networks, information needs to be gathered to one point in time, or sent to the entire network through one point. The location of communication relays and the selection of transfer points in such networks have always been one of the important issues of wireless relays. In conventional base stations or radio communication relays, in order to select suitable communication points, various methods such as theoretical calculations, pure practical measurements, and a combination of theoretical calculations and practical measurements are generally used. In pure theoretical calculations, designers can calculate the approximate signal loss at a specific area distance through the relative distance between maps, the relative height of the transceiver stations, the transmission power of the radio station, the antenna gain, etc., and the known loss formula, and set forwarding points at points where signal loss may cause the data to be unable to be received correctly. Pure practical measurements are road test personnel using signal detection tools to move along the route of the communication relay in the direction away from the signal transmission source. After the signal drops to a specific threshold, the point is selected as the signal forwarding point. The method that combines theory with practice is to first roughly estimate the theoretical attenuation distance, then use a signal meter to measure the signal around the point selected by theoretical calculation, and then select the appropriate threshold to determine the forwarding point.

[0003] In the process of selecting wireless relay points, the selection of existing wireless relay points is generally obtained by road test personnel through actual measurement or theoretical calculation one by one. It requires a lot of manpower and material resources. In the unmanned system communication relay, some existing patents and some methods use theoretical calculations to calculate the theoretical path loss between any two points. And select the relay at the appropriate point of the theoretical path loss to ensure the relay transmission and reception of data.

[0004] For example, the invention patent "Cooperative Control Method of UAV Cluster with Intertwined Tasks, Communications and Topologies" (patent application number: CN201310170260.9) applied by Dai Qionghai and others from Tsinghua University proposed a cooperative control method of UAV cluster with intertwined tasks, communications and topologies, including the following steps: establishing the correlation function of the tasks, communications and topologies of the UAVs; establishing the interactive communication number strategy set, the task information matrix strategy set and the relative distance strategy set between interactive communication UAVs, and determining the winning matrix and related constraints; obtaining the interactive communication number of UAVs, the task information transmission amount and the relative distance between UAVs that meet the preset requirements in the interactive communication number strategy set, the task information matrix strategy set and the relative distance between UAVs; and controlling the UAV cluster according to the interactive communication number of UAVs, the task information transmission amount and the relative distance between UAVs that meet the preset requirements. This patent application constructs the communication capability between UAVs through theoretical calculations;

[0005] In fact, in traditional wireless relay networks, if the communication effect is not ideal after the relay point is determined according to the theoretical model, the position of the relay point can be adjusted. Since the relay points are all on the ground, this adjustment may not be troublesome. However, in the aerial drone relay network, especially in the case of multiple relays, the adjustment of a relay point may affect the relay topology of the entire network. For example, after a relay point closest to the source point is deployed according to the theoretical calculation method, there is environmental interference around it, resulting in poor communication effect. At this time, the position of the relay point needs to be moved closer to the source point. However, after the point moves, the distance between it and the next relay point will be lengthened, which will cause the second hop relay to deteriorate. If the position of the second hop is adjusted further, it will affect the communication quality between the second hop and the third hop, and so on.

[0006] The patent "Method and system for acquiring UAV formation communication topology based on minimum tree graph" (patent application number CN201610384144.0) applied by Luo He and others from Hefei University of Technology provides a method and system for acquiring UAV formation communication topology based on minimum tree graph, the method comprising: constructing a formation communication graph according to the formation of a preset formation; calculating the minimum tree graph of the formation communication graph; judging whether the UAV to which the root node of the minimum tree graph belongs can serve as the formation leader, and obtaining the optimal communication topology according to the judgment result. The method constructs a formation communication graph according to the formation of a preset formation, and by calculating the minimum tree graph of the formation communication graph, judges whether the UAV to which the root node of the minimum tree graph belongs can serve as the formation leader, and obtains the optimal communication topology according to the judgment result, thereby ensuring that the overall communication cost is minimized during the formation maintenance process. The method generates a tree graph based on the formation. The formation is known, and the topology is not specially designed for communication relay;

[0007] The academic paper "Path Planning and Communication Optimization in UAV Collaborative Relay Process" (System Engineering and Electronic Technology) by Fu Xiaowei et al. focuses on theoretical research and arranges nodes according to the theoretical communication distance of the nodes; the academic paper "UAV Relay Layout Optimization Software Based on Google Earth" (Modern Electronic Technology) by Li Yanwen et al. also conducts theoretical modeling of UAV relay, derives the coverage area of ​​the UAV at a certain altitude, and designs static optimization configuration forwarding points.

[0008] The above all belong to the deployment of points using theoretical models. The actual communication effect of the deployed drones is difficult to determine, and there is no specific method for how multiple drones can cooperate with each other to achieve relay. In traditional wireless relay, if the points are deployed according to the theoretical model, if the deployment effect is not ideal, the deployment position can be adjusted. Since the relay points are all on the ground, the adjustment may not be troublesome. In the case of unmanned aerial relay, especially under multi-point relay conditions, the adjustment of one point may affect the relay topology of the entire network.

[0009] For example, after a relay point closest to the source point is arranged according to the theoretical value, the communication effect is poor due to some environmental interference around it. At this time, the relay point needs to be moved closer to the source point. After the point is moved, the distance between the point and the next relay point will be lengthened, resulting in the deterioration of the second-hop relay. Adjusting the position of the second hop will affect the distance between the second and third hops, etc. Reasonable settings are tedious and time-consuming processes.

[0010] In the process of selecting wireless relay points, the selection of existing wireless relay points is generally obtained by road test personnel through actual measurement or theoretical calculation one by one, which requires a lot of manpower and material resources.

[0011] In the process of UAV swarm communication relay, what kind of relationship should exist between multiple UAVs? This relationship is not mentioned in any existing solutions.

[0012] If multiple drones can coordinate with each other, automatically move toward or away from each other at a specific beat, and deliver data when they move toward each other very close, the nodes moving away from each other will not affect the nodes that are communicating. In this way, the relay nodes only need to be deployed according to specific rules and distribution density, without considering whether the deployment points are within the communication range or how the communication effect is. The actual communication is completed through the rhythmic coordinated movement between nodes, which solves the problem of unmanned system deployment in a large area, thereby improving the efficiency of data aggregation or distribution. This is the problem to be solved by the present invention. Summary of the invention

[0013] The present invention provides a convergence and distribution resonant communication method for multiple unmanned systems to cooperate without theoretical calculation of points and without human intervention. According to predetermined procedures and rules, multiple unmanned systems cooperate with each other to successively form a dynamic resonant communication link, thereby completing the periodic convergence and / or distribution of information.

[0014] A convergence distribution resonant communication method for multiple unmanned systems collaboration includes the following steps:

[0015] S1: The relay drone uses the positioning system to calculate the distance from its own position to the convergence point S;

[0016] S2: The relay drone determines which regional node it belongs to based on the distance from itself to the convergence point S, and determines which sub-cycle beat it currently belongs to;

[0017] S3: Each level of cellular nodes completes the data transmission process in one cycle;

[0018] S4: Determine whether the node is accessible; if yes, proceed to the next step; if no, return to step S1 and recalculate;

[0019] S5: Send data until all packets are sent or the safe distance is reached, then return to the original position and end.

[0020] Preferably, each relay drone of the present invention carries a positioning system for determining its own position, and the distance from the relay drone to the convergence point S is calculated through the ability of the relay drones to communicate with each other.

[0021] Preferably, the present invention gives a data area Q and the number of relay drones N that currently complete the data task in the area, and obtains a regional density of Q / N. When uniformly distributed, the distance D between each relay drone node and other adjacent nodes is In area Q, the area Q is divided into one or several levels of cellular nodes according to the distance between each relay drone node and the convergence point S. The radius of each level of cellular nodes is an integer multiple of D.

[0022] Each level of cellular nodes completes the data transmission process within one cycle, and one cycle includes two beats: contraction and relaxation.

[0023] Preferably, in the present invention, during the cycle of completing the data transmission process, the center point of each level of cellular nodes is the location where the relay drone gathers and diffuses information. In the first sub-cycle, the drones located at the vertices of each cellular node perform contraction and relaxation movements toward their center points. By setting the two parameters of the farthest communication distance and the safety distance, the drone begins to gather and diffuse data when it moves to the farthest communication distance range from the convergence point or other adjacent nodes. When it moves to the boundary of the set safety distance range, it stops contracting and expands in the opposite direction until it returns to the vertex position of the regular hexagon where the drone was originally located, completing the data transmission process. In the remaining two sub-cycles, the drone movement mode is basically the same as the first sub-cycle.

[0024] The present invention is actually a method for realizing delay-tolerant intermittent communication of streaming data. The resonant communication method for realizing information aggregation and distribution of unmanned systems proposed in the present invention can cooperate with each other according to predetermined procedures and rules without theoretical calculation and manual intervention, and successively form a dynamic resonant communication link to complete the periodic aggregation and / or distribution of information. This method has the following advantages: first, the relay nodes do not need to communicate with each other to be reachable, so when there are fewer relay nodes in the unmanned system, periodic data delivery in the coverage area can be achieved; second, compared with conventional opportunistic communication, its movement is controlled, and the communication delay and communication reliability are greatly improved; third, the communication equipment it uses can shorten the communication distance to a very short distance, so as not to interfere with the communication between other nodes; fourth, compared with other information aggregation and distribution data transmission methods, its data transmission has stronger continuous data transmission capabilities and less aggregation and distribution data transmission delay; in addition, due to the additional movement and energy consumption overhead, the overall working time of the system is longer; fifth, through resonant information transmission, the movement range of each node (or device) does not need to be too long, and through the relay of multiple nodes, effective information transmission over ultra-long distances can be achieved. Sixth, for data aggregation and diffusion networks, "resonant transmission" brings small end-to-end delay, which is particularly suitable for end-to-end distribution and aggregation of streaming information. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the working status of the cellular node in each sub-cycle. Figure 1 (a) is a schematic diagram of the working status of the cellular node in the first sub-cycle; Figure 1 (b) is a schematic diagram of the working status of the cellular node in the second sub-cycle; Figure 1 (c) is a schematic diagram of the working status of the cellular node in the third sub-cycle.

[0026] Figure 2 This is a schematic diagram of the distribution of unmanned systems in the hexagonal coverage area.

[0027] Figure 3 It corresponds to Figure 2 In the state of the cellular node, when the first sub-cycle shrinks, the cellular where the convergence point S is located "shrinks".

[0028] Figure 4 It corresponds to Figure 2 In the state of the cellular node, the cellular where the convergence point S is located is "relaxed" during the contraction beat of the first sub-cycle.

[0029] Figure 5 It corresponds to Figure 2 In the state of the cellular node, when the second sub-cycle shrinks, the adjacent cells of the convergence point S "shrink".

[0030] Figure 6 It is a flow chart of the convergence and distribution resonant communication method of the present invention. DETAILED DESCRIPTION

[0031] The technical solution of the present invention is described in detail below with reference to the accompanying drawings:

[0032] like Figure 6 As shown, a convergence distribution resonant communication method for multiple unmanned systems collaboration includes the following steps:

[0033] S1: The relay drone uses the positioning system to calculate the distance from its own position to the convergence point S;

[0034] S2: The relay drone determines which regional node it belongs to based on the distance from itself to the convergence point S, and determines which sub-cycle beat it currently belongs to;

[0035] S3: Each level of cellular nodes completes the data transmission process in one cycle;

[0036] S4: Determine whether the node is accessible; if yes, proceed to the next step; if no, return to step S1 and recalculate;

[0037] S5: Send data until all packets are sent or the safe distance is reached, then return to the original position and end.

[0038] The method involved in the present invention involves multiple unmanned systems to realize the distribution and aggregation of streaming information, provided that each unmanned system carries a positioning system such as GPS, Beidou, etc. to determine its own position. At the same time, the unmanned systems also have the ability to communicate with each other, and all nodes have known the location information of the aggregation point S in advance. Given a data aggregation / diffusion area Q and the number of unmanned systems N that currently complete the data aggregation / diffusion task in the area, the unmanned system can calculate the regional density (density is Q / N), and roughly understand the distance D (which can be approximated as) between each unmanned node and other adjacent nodes when uniformly distributed through the density. ). In area Q, each node can be divided into one or more levels according to the distance between the node and the sink point S. Figure 1 Take the four-level partition of the unmanned system as an example, where the first level is nodes with a distance less than or equal to D, and they are in a honeycomb with a radius of D; the second level is nodes greater than D and less than or equal to 2D, the third level is nodes greater than 2D and less than or equal to 3D; the fourth level is nodes greater than 3D and less than or equal to 4D. The present invention divides the movement behavior of the node within one cycle into three sub-cycles, and one sub-cycle includes two beats (contraction and relaxation). The specific periodic motion process of the unmanned system is as follows:

[0039] The first sub-period: Figure 1 As shown in (a), the center point of each regular hexagon is the location where the relay drone gathers and diffuses information. In the first sub-period, the drones at the vertices of the regular hexagons in the 1st-level partition, the 6 vertices of the regular hexagons in the 3rd-level partition, and the 6 vertices of the regular hexagons in the 4th-level partition contract and relax toward their center points. By setting the two parameters of the farthest communication distance and the safety distance, the drone begins to gather and diffuse data when it moves to the farthest communication distance range from the convergence point or other adjacent nodes. When it moves to the boundary of the set safety distance range, it stops contracting and expands in the opposite direction until it returns to the regular hexagon vertex position where the drone was originally located. Since the drone movement time scale is much larger than the data transmission time scale, it is assumed that the drone can complete the data transmission process by starting to transmit data from the farthest communication distance, moving to the safety distance and then returning to the farthest communication distance. In the remaining two sub-periods, the drone movement mode is basically the same as the first sub-period.

[0040] Second sub-period: Figure 1 As shown in (b), in the second sub-cycle, the drones at the 3 regular hexagonal vertices in the 2nd-level partition, the 3 regular hexagonal vertices in the 3rd-level partition, and the 6 regular hexagonal vertices in the 4th-level partition perform contraction and relaxation movements toward their center points.

[0041] The third sub-period: Figure 1As shown in (c), in the third sub-cycle, the drones at the 3 regular hexagonal vertices in the 2nd-level partition, the 3 regular hexagonal vertices in the 3rd-level partition, and the 6 regular hexagonal vertices in the 4th-level partition perform contraction and relaxation movements toward their center points.

[0042] Three sub-cycles complete a large cycle, which can be obtained from Figure 1 It can be seen from the figure that after such a long cycle, the drone can spread the data of the central source node to all drones in the entire area, and at the same time gather the data of drones in the periphery to the central source node. Figure 1 It is deduced that in such an intermittent data diffusion / aggregation process, the integrity and reliability of data transmission can still be guaranteed by using the existing transmission mechanism.

[0043] With S as the center, the nodes can form a variety of shapes. Here, a hexagonal shape is used as an example. Each node in the network knows the density of nodes in the network, and knows which level of area it is in based on the location of its own convergence point. Figure 2 As shown. First, after each unmanned system is turned on, it flies (or moves) to the pre-set position coordinates and stops (rotor drones hover, fixed-wing drones fly around a certain point), and records its own regional level, which is recorded as 1, 2, 3, .... All unmanned systems that achieve coverage are turned on one after another and keep synchronized, and they do not need wireless communication with each other. We take the source point S to the level 1 regional relay node as an example to illustrate.

[0044] The present invention numbers the unmanned systems at the edge of the level 1 area near the convergence point S as U S1 , U S2 ,…U S6 The whole network calculates the beat cycle based on the distance D between the level 1 regional unmanned system and point S. The beat cycle must satisfy the level 1 unmanned system to be able to move to the safe distance of point S within the beat cycle time, and have enough time to obtain the packets sent by the convergence point S or send the packets that the level 1 nodes need to interact to the convergence point S. In a sub-cycle, the unmanned system moves in two beats (i.e., the two movement processes of the drone's contraction and relaxation). Figure 3 As shown, in the first beat (contraction beat), the unmanned system U S1 , U S2 ,…U S6 Move from the edge of the cell to the convergence point S (the cell where the convergence point S is located can be considered to be "shrinking") until the end of the beat or until a safe distance is reached. S1 , U S2 ,…U S6After moving to the farthest communication distance from the convergence point S and reaching the communication range of the convergence point S, they can communicate with each other and send data. The closer the distance, the better the communication effect. They stop at a safe distance between each other. At this time, the convergence point S sends data to all level 1 regional nodes (distribution) or all level 1 regional nodes send data to the convergence point S (aggregation). The receiving data node returns ACK to the sending unit. When the sending node receives ACK, it stops sending data, otherwise it continues to send until it receives ACK from all nodes.

[0045] On the second beat (diastolic beat, such as Figure 4 ), the cell where the convergence point S is located begins to "expand", U S1 , U S2 ,…U S6 , the six unmanned systems return to their original positions respectively, completing a complete "contraction and relaxation" movement.

[0046] At this time, the system enters the first beat of the second sub-cycle. Figure 5 As shown, the three neighboring cells of the aggregation point S begin to "shrink", thereby completing the diffusion of data from the inside to the outside (data distribution) or the collection of data from the outside to the inside (data aggregation). At this time, the primary node ends the data distribution after receiving the ACK returned by the secondary node, or the secondary node ends the data aggregation after receiving the ACK returned by the primary node.

[0047] Since the data broadcast from or collected to the convergence point S is transferred by one or more relay nodes through periodic resonant motion, this data transmission method is called "resonant communication", and its motion pattern is similar to the blood circulation caused by the contraction and relaxation of the heart. Each unmanned system node does not need to move a long distance, it only needs to be able to contact the next hop. At the same time, the number of intermediate forwarding nodes can be more or less. When there are fewer nodes, the resonant motion distance of the unmanned system is large, and when there are more nodes, the resonant motion distance of the unmanned system is small. Therefore, the number of unmanned system nodes required for this data transmission method can be more or less, and the ability to adapt to the scene is relatively strong. In "resonant communication", data is continuously transmitted from the source point to the destination point in the form of a stream, ensuring the freshness of the data. Since the energy consumption of the unmanned system when flying or moving mainly depends on the acceleration or resistance, when the acceleration does not change much, the additional energy consumption of the unmanned system can be ignored; on the other hand, since adjacent nodes can move very close before communicating, "resonant communication" reduces the communication time and energy consumption on the one hand, and reduces interference to other non-adjacent nodes on the other hand.

[0048] This embodiment provides a schematic diagram of building data aggregation and distribution for unmanned systems. Without loss of generality, the unmanned system in this diagram is represented by a drone. Figure 2 . Figure 2 S is the source point where data aggregation or distribution is required. Assuming that the obstruction between the aggregation point S and the destination node does not exceed M meters (assuming M is 100 meters), the lift-off height of all drones can be set to M meters. This can avoid the influence of all obstructions. Assuming that there are N drones around the aggregation point S as data aggregation and distribution nodes (assuming there are 10 drones), in the schematic diagram of this case, 6 drones can be distributed in a honeycomb shape, and N-6 drones can be arranged in a honeycomb shape at the farthest data collection point. Without loss of generality, assuming that the cell radius is R meters (assuming R = 50 meters), the farthest communication distance of the drone communication equipment we choose is D meters (assuming D = 20 meters), and the safety distance is set to Q meters (assuming Q = 1 meter). Then, the 6 drones at the edge of the cell are all level 1 regional nodes; the other cell is a level 2 regional node. The aggregation point S has data to send. Assume that the drone's flight speed is V meters / second (assuming V = 5 meters / second). In the contraction rhythm of the first sub-cycle, the node in area 1 “contracts” and the six drones converge toward the convergence point S at a speed of 5 m / s. After 6 seconds, the 6 drones move to the farthest communication distance D from the convergence point S and can communicate with point S. After 9.8 seconds, the drones approach the safe distance. To prevent collisions between drones and between drones and the convergence point S, the drones stop flying and continue to send data when they reach the safe distance until the data is sent. The first-level node sends an ACK back to the convergence point S. The convergence point S stops sending after receiving the ACK, otherwise it will time out and retransmit. When the data is transmitted, the six nodes in the first-level area enter the relaxation rhythm and return to the edge of the cell. After 19.6 seconds, it returns to the starting position of the 6 nodes. The first sub-cycle ends. At this time, it enters the second sub-cycle contraction rhythm. The second sub-cycle is also Seconds, including contraction beats seconds, and the diastolic beat is also The two first-level regional nodes on the right and other second-level regional nodes begin to "shrink" toward the center of the second-level cell. After a few seconds, the data will be exchanged. After a few seconds, the "contraction beat" ends, and the secondary node sends an ACK to the primary node, otherwise it will retransmit. After the data transmission is completed, it starts to enter the "relaxation beat" and returns until After a few seconds, the original positioning position is restored. The second sub-cycle ends.

[0049] The beat can be set to 10 seconds, and each sub-cycle contains two beats (systolic beat and diastolic beat). Since there are only two cells in this example, two sub-cycles (20 seconds each) can complete the diffusion or data aggregation task. During the flight, the data is aggregated (or distributed). At this time, it enters the first sub-cycle. The entire process can be completed automatically by the drone according to the pre-set procedures and processes without human intervention.

[0050] The unmanned system topology used to achieve data aggregation can be a quadrilateral, octagon or other topology that can cover the area. According to a similar multiple beat or a further refinement of the beat just described, the final result is that the data of the aggregation point S is periodically distributed or aggregated. Or multiple points are partitioned by area to achieve periodic distribution and aggregation of data. The minimum distance between adjacent nodes may not be a safe distance, but may be other suitable communication distances, or some method may be used to ensure that the amount of forwarded data does not exceed the time of the opposite movement.

[0051] The motion synchronization between multiple drones can be achieved through Beidou short messages, or through a time-sharing agreement mechanism under a unified agreement or with the cooperation of other time-unified devices (such as built-in high-stability clocks such as rubidium clocks).

[0052] In addition to drones (rotor-wing and fixed-wing), other information transmission devices with relatively longer communication distances, such as unmanned boats, unmanned submarines, and aircraft in space, can adopt similar resonant communication forwarding mechanisms. While ensuring that other nodes do not interfere with communication nodes, periodic information aggregation and distribution can be achieved through rhythmic repeated data interaction, and communication interference to other nodes can be reduced.

Claims

1. A convergence distribution resonant communication method for multiple unmanned systems, characterized in that The steps include: Step S1: The relay drone uses the positioning system to calculate the distance from its own position to the convergence point S; The specific process is as follows: each relay UAV carries a positioning system to determine its own position, and through the ability of relay UAVs to communicate with each other, the distance from the relay UAV to the convergence point S is calculated; Step S2: The relay drone determines which regional node it belongs to based on the distance from itself to the convergence point S, and determines which sub-cycle beat it currently belongs to; The specific process includes: Given a data area Q and the number of relay drones N that currently complete the data task in the area, the regional density is Q / N. When evenly distributed, the distance D between each relay drone node and other adjacent nodes is In area Q, the area Q is divided into one or several levels of cellular nodes according to the distance between each relay drone node and the convergence point S. The radius of each level of cellular nodes is an integer multiple of D. Each level of cellular nodes completes the data transmission process within one cycle, and one cycle includes two beats: contraction and relaxation; Step S3: Each level of cellular nodes completes the data transmission process in one cycle; Step S4: Determine whether the node is accessible; if yes, proceed to the next step; if no, return to step S1 and recalculate; S5: Send data until all packets are sent or the safe distance is reached, then return to the original position and end.

2. The convergence distribution resonant communication method for multi-unmanned system collaboration according to claim 1 is characterized in that During the above-mentioned cycle of completing the data transmission process, the center point of each level of cellular node is the position where the relay drone gathers and diffuses information. In the first sub-cycle, the drones located at the vertices of each cellular node perform contraction and relaxation movements toward their center points. By setting the two parameters of the farthest communication distance and the safety distance, the drone begins to gather and diffuse data when it moves to within the farthest communication distance range from the convergence point or other adjacent nodes. When it moves to the boundary of the set safety distance range, it stops contracting and expands in the opposite direction until it returns to the vertex position of the regular hexagon where the drone was originally located, completing the data transmission process. In the remaining two sub-cycles, the drone movement mode is basically the same as the first sub-cycle.

Images