Distributed swarm motion control method for UAVs in communication-denied environments

Abstract

This application relates to a method for controlling the motion of a distributed swarm of drones in a communication-denied environment. The method includes generating local environmental information concentrations about mission targets and threat zones in the current environment based on a gene regulatory network; determining, based on the local environmental information concentrations, a group velocity action vector for maintaining the swarm's morphology, an evasion velocity action vector for avoiding threat zones, a tracking velocity action vector for tracking the mission target, and a coordinated observation velocity action vector for neighboring drones; and designing a controller for the drones based on their motion constraints. The controller then performs distributed swarm motion control. This method enables distributed swarm control of drones.

Description

Technical Field

[0001] The present application relates to the technical field of drone cluster motion control, and in particular to a method for distributed drone cluster motion control in a communication denial environment. Background Art

[0002] With the development of drone technology, drones are being applied in various scenarios, such as emergency network support, interference suppression, and navigation deception. Take drone emergency support as an example. When emergency support is needed in a certain area, a drone swarm is required to reach the target area to carry out the emergency support mission. Since the emergency event may not have been resolved, each target area presents a potential threat. The current conventional technology is for a control center to communicate with the drone swarm and issue task execution instructions to the drone swarm to carry out the emergency support. This method is impossible if communication is lost. In addition, if a single drone cannot maintain a swarm formation, the advantages of swarm emergency support cannot be realized. Currently, there are also drone swarm designs based on intelligent agents, but they cannot meet the needs of complex scenarios. Summary of the Invention

[0003] Based on this, it is necessary to provide a distributed cluster motion control method for drones in a communication denial environment to address the above technical problems.

[0004] A method for controlling the motion of a distributed swarm of unmanned aerial vehicles (UAVs) in a communication-denied environment, the method comprising:

[0005] Generate local environmental information concentration about mission targets and threat areas in the current environment based on gene regulatory networks;

[0006] Determine, based on the local environmental information concentration, a group speed action vector for the drones to maintain their group form, an avoidance speed action vector for the drones to avoid the threat area, a tracking speed action vector for the drones to track the mission target, and a friendly drone observation and coordination speed action vector;

[0007] Design a controller for the drone based on the vector sum of the group velocity action vector, the avoidance velocity action vector for the drone to avoid the threat area, the tracking velocity action vector for the drone to track the mission target, and the observation and coordination velocity action vector of the neighboring drone, as well as the motion constraints of the drone;

[0008] Use the designed controller to control the distributed cluster motion of drones.

[0009] In one embodiment, the method further includes setting the group velocity action vector of the drones to maintain the group shape to:

[0010] V e =k e ·(Cd -C m )·P e

[0011] Among them, V e represents the group velocity action vector, k e represents an adjustable parameter, C d Indicates the concentration value of the drone position in the grid with the preset range centered on the drone’s location, C m Indicates the minimum concentration in the grid, P e The direction vector indicating the grid where the drone is located pointing to the grid with the minimum concentration.

[0012] In one embodiment, the method further includes: obtaining a speed attenuation curve of the UAV avoiding the threat area:

[0013]

[0014] Where r is the distance between the drone and the expected stopping point, a is the acceleration parameter, and p is the linear gain;

[0015] According to the speed attenuation curve, the evasion speed action vector of the UAV to evade the threat area is obtained as:

[0016]

[0017] Among them, r ds is the distance from the UAV to its nearest obstacle, r shill is the distance from the drone to its desired obstacle avoidance stopping point, p shill and a shill is the linear gain and acceleration parameter, V s Represents the velocity vector of the virtual agent, V ds Represents the components of the tracking velocity action vector.

[0018] In one embodiment, the method further includes setting the speed action vector of the drone tracking the mission target to be:

[0019]

[0020] P t Represents the direction vector pointing from the UAV to the corresponding mission target;

[0021] Set the velocity action vector of the drone to avoid collision with other mission targets:

[0022]

[0023] Among them, P o Represents the direction vector from the UAV to other UAVs in the grid;

[0024] According to the speed action vector of the drone tracking the corresponding mission target and the speed action vector of the drone avoiding collision with other mission targets, the tracking speed action vector of the drone tracking the mission target is calculated as:

[0025] V d =V ds +V do

[0026] In one embodiment, the method further includes: setting a repulsive speed effect of a friendly UAV based on a position threat according to the position of the friendly UAV within the grid;

[0027] According to the speed of the friendly drones in the grid, set the repulsive speed effect of the friendly drones based on speed threats;

[0028] Set the speed alignment of neighboring drones based on the average speed vector of neighboring drones within the grid;

[0029] According to the repulsive speed effect of friendly UAVs based on position threat, the repulsive speed effect of friendly UAVs based on speed threat and the speed alignment effect of friendly UAVs, the observation cooperative speed effect vector of friendly UAVs is obtained.

[0030] In one embodiment, the method further includes setting the repulsion speed of the position-based threat-based friendly UAV according to the position of the friendly UAV in the grid:

[0031]

[0032] Among them, V np represents the repulsive velocity effect, r ij Represents drones i and r rep The distance between other drones j in the range, k np The speed influence coefficient of the UAV considering the positions of other friendly UAVs, P n represents the direction vector of UAV i towards its neighbor UAV j.

[0033] In one embodiment, the method further includes setting the repulsive speed of the friendly UAV based on the speed threat to be:

[0034] According to the repulsive speed effect of the friendly UAV based on speed threat and the speed alignment effect of the friendly UAV, the UAV is considered based on the above behavior. rep The combined speed effect V generated by the speed of other drones within the range nv for:

[0035]

[0036]

[0037] Among them, k nf k is the velocity alignment coefficient of the neighboring UAV, nv The speed influence coefficient of the UAV considering the speed of other friendly UAVs, V j is the speed of UAV j, r ij Represents drones i and r rep The distance between other drones j in the range, N d Indicates drone r rep the number of drones in range;

[0038] According to the repulsive speed effect of the friendly UAV based on the position threat and based on the above behavior considerations rep The comprehensive speed effect produced by the speed of other UAVs within the range is obtained, and the coordinated speed effect vector of the friendly UAV observation is obtained as follows:

[0039] V n =V np +V nv

[0040] Among them, V n Represents the coordinated velocity action vector of the friendly UAV observation.

[0041] In one embodiment, the method further includes designing a controller of the drone based on the sum of the group velocity action vector, the avoidance velocity action vector of the drone avoiding the threat area, the tracking velocity action vector of the drone tracking the mission target, and the observation and cooperation velocity action vector of the neighboring drone, as well as the motion constraints of the drone:

[0042] V=V l ·R+(1-R)·(V e +V t +V d +V n )

[0043]

[0044] Among them, R is the inertial motion coefficient of the UAV, V max V is the maximum speed of the drone under current conditions, l The speed of the drone at the last moment calculated by the controller.

[0045] A distributed cluster motion control device for unmanned aerial vehicles in a communication-denied environment, the device comprising:

[0046] A concentration calculation module is used to generate local environmental information concentrations about mission targets and threat areas in the current environment based on the gene regulatory network;

[0047] A speed action vector calculation module is used to determine the group speed action vector for the drones to maintain the group form, the avoidance speed action vector for the drones to avoid the threat area, the tracking speed action vector for the drones to track the mission target, and the observation and coordination speed action vector of the friendly drones based on the local environmental information concentration;

[0048] a controller design module for designing a controller for the UAV based on the vector sum of the group velocity action vector, the avoidance velocity action vector for the UAV to avoid the threat area, the tracking velocity action vector for the UAV to track the mission target, and the observation and coordination velocity action vector of the neighboring UAV, as well as the UAV's motion constraints;

[0049] The control module is used to control the distributed cluster motion of UAVs using a designed controller.

[0050] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0051] Generate local environmental information concentration about mission targets and threat areas in the current environment based on gene regulatory networks;

[0052] Determine, based on the local environmental information concentration, a group speed action vector for the drones to maintain their group form, an avoidance speed action vector for the drones to avoid the threat area, a tracking speed action vector for the drones to track the mission target, and a friendly drone observation and coordination speed action vector;

[0053] Design a controller for the drone based on the vector sum of the group velocity action vector, the avoidance velocity action vector for the drone to avoid the threat area, the tracking velocity action vector for the drone to track the mission target, and the observation and coordination velocity action vector of the neighboring drone, as well as the motion constraints of the drone;

[0054] Use the designed controller to control the distributed cluster motion of drones.

[0055] The above-mentioned method for controlling the distributed swarm motion of drones in a communication-denied environment generates local environmental information concentrations about the mission target and threat zone in the current environment based on a gene regulatory network. Then, based on the local environmental information concentrations, the group velocity action vector for the drones to maintain the swarm morphology, the avoidance velocity action vector for the drones to avoid the threat zone, the tracking velocity action vector for the drones to track the mission target, and the friendly drone observation and coordination velocity action vector are determined. Based on the vector sum of the group velocity action vector, the avoidance velocity action vector for the drones to avoid the threat zone, the tracking velocity action vector for the drones to track the mission target, and the friendly drone observation and coordination velocity action vector, as well as the motion constraints of the drones, the drone controller is designed. The designed controller is used to perform distributed swarm motion control of drones. The present invention can complete the swarm roundup mission when controlling the distributed swarm of drones, and the controller design of a single drone can also enable it to maintain the swarm morphology when performing the mission. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A gene regulatory network concentration mapping curve diagram in one embodiment;

[0057] Figure 2 An environmental information concentration map for mapping a gene regulatory network in one embodiment;

[0058] Figure 3 An environmental information concentration map for mapping a gene regulatory network under local information in one embodiment;

[0059] Figure 4 1 is a flow chart of a method for controlling the motion of a distributed swarm of drones in a communication-denied environment according to one embodiment;

[0060] Figure 5 A curve showing how the speed force exerted by a drone on a captured target and a non-captured target varies with distance in one embodiment;

[0061] Figure 6 A schematic diagram of 12 drones performing distributed cluster movement to counter an enemy target in a static threat zone scenario in one embodiment;

[0062] Figure 7 A schematic diagram of 16 drones performing distributed cluster movement to counter an enemy target in a static threat zone scenario in one embodiment;

[0063] Figure 8 A schematic diagram of 20 drones performing distributed cluster movement to counter an enemy target in a static threat zone scenario in one embodiment;

[0064] Figure 9A schematic diagram of an embodiment of using the GRN method to conduct distributed cluster movement of 20 drones in a dynamic threat zone scenario to counter two targets;

[0065] Figure 10 A schematic diagram of an embodiment of using the AGENT method to conduct distributed cluster movement of 20 drones in a dynamic threat zone scenario to counter two targets;

[0066] Figure 11 A schematic diagram of an embodiment of the present invention using the method to conduct distributed cluster movement of 20 drones in a dynamic threat zone scenario to counter two targets;

[0067] Figure 12 This is a structural block diagram of a distributed cluster motion control device for UAVs in one embodiment;

[0068] Figure 13 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0070] In a scenario where a distributed swarm of drones is used to capture a target drone, the drones in the swarm are equipped with visual sensors, lidar, and other sensing equipment. Using communication equipment without interference, the drones can obtain speed and location information of neighboring drones within a certain range using their onboard equipment.

[0071] Assuming that communications are not interfered with, the drone can obtain the target's position, velocity, and threat zone location in real time. If communications are cut off after performing a countermeasure mission for a period of time, the drone can rely on its onboard sensors, such as visual sensors and lidar, to obtain the target's position, velocity, and the location of the threat zone within a certain range (sensor detection range).

[0072] Population morphological planning is achieved through gene regulatory networks. Gene regulation is the mechanism by which organisms control gene expression. Certain microorganisms exhibit a high degree of adaptability to environmental conditions during development, being able to rapidly adjust the expression levels of different genes in response to environmental changes. This study focuses on how microorganisms adapt to their environment by altering their metabolism through gene regulation. Adaptability is the ability of a biological system to return to its original state after responding to external perturbations. It is also a key function of gene regulatory networks, reflecting the ability of an organism to survive in a changing environment. Numerous examples of adaptability exist in biological systems, such as bacterial chemotaxis, bacterial anabolism, and transmembrane reactions in yeast.

[0073] In recent years, the use of gene regulatory networks and morphological gradients has been proposed for morphological planning in swarm robotic systems. Microbial gene regulatory networks regulate protein production based on environmental information. Changes in protein concentration, in turn, influence their response to environmental information, further impacting microbial behavior. This study establishes a mapping between microorganisms and drones, applying the gene expression mechanisms of biological morphogenesis to the morphogenesis of swarm drones. Using gene regulatory network mechanisms, drones establish a concentration field that comprehensively captures information about countermeasure targets and threat zones in their environment.

[0074] Commonly used mathematical models of gene regulatory networks include directed graph models, Boolean network models, Bayesian network models, differential equations, difference equation models, etc. For example, Equation (1) indicates that the concentration p(t+1) at time step t+1 is the concentration p at time step t. j The Laplace operator (or second-order spatial derivative) of (t) indicates that the temporal variation of concentration is related to the "curvature" of its spatial distribution. Equation (2) is the definition of the Laplace operator, which represents the sum of the second-order spatial derivatives in the x and y directions and reflects the spatial distribution of concentration. These two formulas can also be viewed as a simulation of the diffusion of protein concentration in biological systems.

[0075] p j (t+1)=▽ 2 p j (t), (1)

[0076]

[0077] However, in a robotic system, the robot may establish a continuously updated decision-making mechanism to iteratively move step by step. Therefore, a differential equation is used to describe this characteristic, as shown in Equations (3) and (4).

[0078]

[0079] Where x(t) and y(t) are the functions of the two distance variables from the UAV to the target or threat area changing with time t. v is the concentration diffusion factor, which can adjust the mapping relationship between the distance interval and the concentration value. j (t) represents the protein concentration generated by the location information of the jth target or threat area, N t is the total number of detected target or threat area grids. Equation (4) calculates the number of grids from j = 1 to j = N t , the comprehensive protein concentration produced by all detected targets or threat areas at the current location of the drone.

[0080] After processing the threat zone and target location information, the drone constructs concentration information fields C1, C2, and C3 on a map. The concentration information mapping mechanism for gene regulatory networks can take various forms. The calculation methods for C1, C2, and C3 vary across different gene regulatory networks. C1 and C2 process the inputs of target location information and threat zone location information, respectively. C3 is the coupled integrated concentration information field, which can also be formalized as a concentration information map for the drone. β1, β2, and β3 are the thresholds of the sigmoid function. Adjusting β adjusts the interval range generated by the entire concentration field. k is the concentration difference adjustment coefficient. A larger k value results in a more pronounced concentration difference. Figure 1 The mapping curve of concentration value and distance under different k and β parameters of the sigmoid function is shown.

[0081] C1(t+1)=sig(p,β1,k), (5)

[0082] C2(t+1)=1-sig(p,β2,k), (6)

[0083] C3(t+1)=sig(C1+C2,β3,k), (7)

[0084]

[0085] Through the above mechanism, the UAV maps the threat area and target location information into a comprehensive concentration information field such as Figure 2 In order to verify the robustness and actual deployment capability of the algorithm, this study uses a local information concentration field. That is, the drone only has a concentration information map within a certain range, such as Figure 3 shown.

[0086] After the UAV generates a local concentration information map based on the threat zone location and the target location, it samples the concentration information map at a certain concentration value, and the closed figure (convex) formed is the target group shape.

[0087] In one embodiment, Figure 4 As shown, a method for controlling the motion of a distributed swarm of drones in a communication-denied environment is provided, comprising the following steps:

[0088] Step 402 : generating local environmental information concentration about the mission target and threat area in the current environment based on the gene regulatory network.

[0089] Step 404, based on the local environmental information concentration, determines the group speed action vector for the drones to maintain the group form, the avoidance speed action vector for the drones to avoid the threat area, the tracking speed action vector for the drones to track the mission target, and the friendly drone observation and coordination speed action vector.

[0090] Step 406, designing a controller for the UAV based on the vector sum of the group velocity action vector, the avoidance velocity action vector for the UAV to avoid the threat area, the tracking velocity action vector for the UAV to track the mission target, and the observation coordination velocity action vector of the friendly UAV, as well as the UAV's motion constraints.

[0091] Step 408: Use the designed controller to perform distributed cluster motion control of UAVs.

[0092] In the above-mentioned method for controlling the motion of a distributed swarm of drones in a communication-denied environment, a distributed finite state machine is designed based on the different target decision-making scenarios that the drone may face, and the states and jump conditions of the finite state machine are set. In addition, a distributed game decision neural network is constructed to achieve optimal target selection when considering different decision variables under the local information of the drone. The network realizes adaptive adjustment of the weight coefficients of each decision variable under local information through the historical data of the drone's environmental information and real-time observations. That is, the drone adjusts parameters while flying. The present invention realizes the autonomous decision-making and selection of mission targets by drones based on the surrounding local information in a communication-denied environment. It can be flexibly applied to a variety of drone decision-making mission target selection mission scenarios and has the practicality of actual machine deployment.

[0093] In one embodiment, the group velocity action vector for the drones to maintain the group shape is set to:

[0094] V e =k e ·(C d -C m )·P e

[0095] Among them, V e represents the group velocity action vector, k e represents an adjustable parameter, C d Indicates the concentration value of the drone position in the grid with the preset range centered on the drone’s location, C m Indicates the minimum concentration in the grid, P e The direction vector indicating the grid where the drone is located pointing to the grid with the minimum concentration.

[0096] Specifically, in the present invention, the drone calculates the concentration value of a 5*5 grid around its location as the center. The concentration value of the grid where the drone is located is C d The minimum concentration of the 5*5 grid is C m , P e k is the direction vector of the grid where the drone is located pointing to the grid with the minimum concentration around the drone, e As an adjustable parameter, the speed effect V is calculated e The drone acts on the speed V eIt can achieve grid motion with fewer friendly drones in a group form.

[0097] In one embodiment, the speed attenuation curve of the drone avoiding the threat area is obtained as follows:

[0098]

[0099] Where r is the distance between the drone and the expected stopping point, a is the acceleration parameter, and p is the linear gain;

[0100] According to the speed attenuation curve, the evasion speed action vector of the UAV to avoid the threat area is obtained as follows:

[0101]

[0102] Among them, r ds is the distance from the UAV to its nearest obstacle, r shill is the distance from the drone to its desired obstacle avoidance stopping point, p shill and a shill is the linear gain and acceleration parameter, V s Represents the velocity vector of the virtual agent, V ds Represents the components of the tracking velocity action vector.

[0103] In this embodiment, the drone needs to avoid a threat zone or obstacle. In a strong confrontation environment, this threat zone may be dynamic. Moreover, when a distributed autonomous drone cannot communicate, its perception range may be limited if it relies solely on recorded sensors. Therefore, a speed function V is designed for the drone to avoid the threat zone under local information. t D(r,a,p) is the speed attenuation curve of the drone avoiding the threat area

[14] Where r is the distance between the drone and the expected stopping point, a is the acceleration parameter, and p is the linear gain, which also determines the intersection between the different deceleration phases of the D(r,a,p) function. ds is the distance from the UAV to its nearest obstacle, r shill is the distance from the drone to its desired obstacle avoidance stopping point, p shill and a shill Linear gain and acceleration parameters. Assume that there is a virtual agent at the nearest obstacle point of the drone, V s is the velocity vector of the virtual agent, which is perpendicular to the obstacle boundary and points to the field. Its size is an adjustable parameter.

[0104] In one embodiment, the speed action vector of the drone tracking the corresponding mission target is set to:

[0105]

[0106] P t Indicates the direction vector from the UAV to the corresponding mission target; the speed action vector of the UAV to avoid collision with other mission targets is set as:

[0107]

[0108] Among them, P o Represents the direction vector pointing from the UAV to other UAVs in the grid; based on the speed action vector of the UAV tracking the corresponding mission target and the speed action vector of the UAV avoiding collision with other mission targets, the tracking speed action vector of the UAV tracking mission target is calculated as:

[0109] V d =V ds +V do

[0110] In this embodiment, the drone needs to get close to the target during the process of capturing the drone target, but it cannot be too close to the target. In order to capture the target, the drone cannot collide with the target. Figure 5 The curve depicts three types of drone behaviors: when the drone is too close to the target, it accelerates (the rate of change / acceleration becomes larger and larger as it moves away from the target), moving away from the target to be countered; when the drone is too far away from the target, it accelerates to track the target to be countered at an appropriate speed; when the drone is at a moderate distance from the enemy target, the speed effect term approaches zero.

[0111] In addition, while avoiding collisions with the counter-targets of their choice, drones must also avoid collisions with other enemy drones.

[0112] In one embodiment, according to the position of the friendly UAV in the grid, the repulsive speed effect of the friendly UAV based on the position threat is set; according to the speed of the friendly UAV in the grid, the repulsive speed effect of the friendly UAV based on the speed threat is set; according to the average value of the speed vectors of the friendly UAVs in the grid, the speed alignment effect of the friendly UAVs is set; according to the repulsive speed effect of the friendly UAV based on the position threat, the repulsive speed effect of the friendly UAV based on the speed threat and the speed alignment effect of the friendly UAVs, the friendly UAV observation collaborative speed effect vector is obtained.

[0113] In one embodiment, the repulsion speed of the position-based threat friendly drone is set based on the position of the friendly drone within the grid:

[0114]

[0115]

[0116] Among them, V np represents the repulsive velocity effect, r ij Represents drones i and r rep The distance between other drones j in the range, k np The speed influence coefficient of the UAV considering the positions of other friendly UAVs, P n represents the direction vector of UAV i towards its neighbor UAV j.

[0117] In addition, according to the speed of the friendly UAV in the grid, the repulsive speed effect of the friendly UAV based on speed threat is set as follows: According to the repulsive speed effect of the friendly UAV based on speed threat and the speed alignment effect of the friendly UAV, the UAV is considered based on the above behavior. rep The combined speed effect V generated by the speed of other drones within the range nv for:

[0118]

[0119] Among them, k nf k is the velocity alignment coefficient of the neighboring UAV, nv The speed influence coefficient of the UAV considering the speed of other friendly UAVs, V j is the speed of UAV j, r ij Represents drones i and r rep The distance between other drones j in the range, N d Indicates drone r rep the number of drones in range;

[0120] According to the repulsive speed effect of the friendly UAV based on the position threat and based on the above behavior considerations rep The comprehensive speed effect produced by the speed of other UAVs within the range is obtained, and the coordinated speed effect vector of the friendly UAV observation is obtained as follows:

[0121] V n =V np +V nv

[0122] Among them, V n Represents the coordinated velocity action vector of the friendly UAV observation.

[0123] In one embodiment, based on the vector sum of the group velocity action vector, the avoidance velocity action vector of the UAV avoiding the threat area, the tracking velocity action vector of the UAV tracking the mission target, and the observation and coordination velocity action vector of the neighboring UAV, as well as the UAV's motion constraints, the UAV controller is designed as follows:

[0124] V=V l ·R+(1-R)·(Ve +V t +V d +V n )

[0125]

[0126] Among them, R is the inertial motion coefficient of the UAV, V max V is the maximum speed of the drone under current conditions, l The speed of the drone at the last moment calculated by the controller.

[0127] In summary, the present invention designs a simulation experiment of UAV distributed cluster in countering illegal flying. In order to verify the robustness of the method proposed in this study, this study designs a single target counterattack experiment with different group sizes in a static threat zone environment ( Figure 6 、 Figure 7 and Figure 8 In order to be closer to the actual environment, the present invention designs a dynamic threat zone multi-target countermeasure experiment. In order to verify the superiority of the algorithm, the present invention applies the control method of the present invention ( Figure 9 ) and the GRN method ( Figure 10 ), and the AGENT method ( Figure 11 ) were subjected to comparative tests.

[0128] from Figure 6 、 Figure 7 、 Figure 8 It can be seen that when the group sizes are 12, 16, and 20 respectively, the distributed countermeasures against a single black flying drone scenario using the method of the present invention can achieve good tracking of the countermeasure target through a complex static threat area.

[0129] from Figure 9 、 Figure 10 、 Figure 11 It can be seen that when the group size is 20, the distributed countermeasure scenario of two targets deployed by the method of the present invention can achieve good tracking and countermeasure behavior of two targets through a complex dynamic threat area, and can also produce a distributed and uniform distribution effect.

[0130] It should be understood that although Figure 4 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 4At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0131] In one embodiment, Figure 12 As shown, a distributed decision-making device for multiple targets of a UAV swarm in a communication denial environment is provided, comprising: a concentration calculation module 1202, a velocity action vector calculation module 1204, a controller design module 1206, and a decision output module 1208, wherein:

[0132] A framework construction module 1202 is used to construct a distributed decision-making finite state machine and a game decision neural network for drones in a drone cluster; the states of the drones in the distributed decision-making finite state machine include: decision keeper, normal decision maker, entangled decision maker, and lost decision maker;

[0133] The parameter adjustment module 1204 is used to adjust the parameters of the game decision neural network in real time in the current decision round based on the decision data stored in the previous decision round by the drone and the real-time data currently acquired;

[0134] The state jump logic setting module 1206 is used to set the state jump rules in the distributed decision finite state machine; the state jump rules include: if the state of the distributed decision finite state machine of the drone is a normal decision maker, and the distance between the drone and any mission target is judged to be greater than the drone decision distance threshold, then the state of the drone is judged to jump to a decision holder; if the short-term decision jump of the drone is judged to be greater than the short-term decision memory capacity of the drone multiplied by the short-term decision tolerance jump rate of the drone, then the state of the drone is judged to jump to a lost person; if the distance of the drone that jumps to a decision holder from any mission target is less than the decision distance Threshold, the state of the decision holder's drone jumps to a normal decision maker. If the current lost state bit of the decision holder's drone is 1, the state of the decision holder's drone jumps to a lost one. If the distance of the drone that jumps to the entangled one from any mission target is greater than the decision distance threshold, the state of the entangled one's drone jumps to a decision holder. If the distance of the drone that jumps to the entangled one from all mission targets increases or decreases, the state of the entangled one's drone jumps to a lost one. If the number of steps counted by the drone that jumps to the lost one reaches the threshold, the state of the lost one's drone jumps to a normal decision maker.

[0135] The decision output module 1208 is used for the drones in the normal decision maker and entangled states to output decisions through the game decision neural network after real-time parameter adjustment. The drone in the lost state determines the optimal mission target of the lost state as the decision output by comparing the number of mission targets and the number of drones within a preset distance. The drone in the decision holder state maintains the decision of the previous decision round.

[0136] In one embodiment, the group velocity action vector for the drones to maintain the group shape is set to:

[0137] V e =k e ·(C d -C m )·P e

[0138] Among them, V e represents the group velocity action vector, k e represents an adjustable parameter, C d Indicates the concentration value of the drone position in the grid with the preset range centered on the drone’s location, C m Indicates the minimum concentration in the grid, P e The direction vector indicating the grid where the drone is located pointing to the grid with the minimum concentration.

[0139] In one embodiment, the speed attenuation curve of the drone avoiding the threat area is obtained as follows:

[0140]

[0141] Where r is the distance between the drone and the expected stopping point, a is the acceleration parameter, and p is the linear gain;

[0142] According to the speed attenuation curve, the evasion speed action vector of the UAV to evade the threat area is obtained as:

[0143]

[0144] Among them, r ds is the distance from the UAV to its nearest obstacle, r shill is the distance from the drone to its desired obstacle avoidance stopping point, p shill and a shill is the linear gain and acceleration parameter, V s Represents the velocity vector of the virtual agent, V ds Represents the components of the tracking velocity action vector.

[0145] In one embodiment, the velocity action vector of the drone tracking the mission target is set to:

[0146]

[0147] P t Represents the direction vector pointing from the UAV to the corresponding mission target;

[0148] Set the velocity action vector of the drone to avoid collision with other mission targets:

[0149]

[0150] Among them, P o Represents the direction vector from the UAV to other UAVs in the grid;

[0151] According to the speed action vector of the drone tracking the corresponding mission target and the speed action vector of the drone avoiding collision with other mission targets, the tracking speed action vector of the drone tracking the mission target is calculated as:

[0152] V d =V ds +V do

[0153] In one embodiment, a repulsive speed effect of a friendly UAV based on a position threat is set according to the position of the friendly UAV within the grid;

[0154] According to the speed of the friendly drones in the grid, set the repulsive speed effect of the friendly drones based on speed threats;

[0155] Set the speed alignment of neighboring drones based on the average speed vector of neighboring drones within the grid;

[0156] According to the repulsive speed effect of friendly UAVs based on position threat, the repulsive speed effect of friendly UAVs based on speed threat and the speed alignment effect of friendly UAVs, the observation cooperative speed effect vector of friendly UAVs is obtained.

[0157] In one embodiment, the repulsion speed of the position-based threat friendly drone is set based on the position of the friendly drone within the grid:

[0158]

[0159] Among them, V np represents the repulsive velocity effect, r ij Represents drones i and r rep The distance between other drones j in the range, k np The speed influence coefficient of the UAV considering the positions of other friendly UAVs, P n represents the direction vector of UAV i towards its neighbor UAV j.

[0160] In one embodiment, based on the speed threat-based repulsive speed effect of the neighboring UAV and the speed alignment effect of the neighboring UAV, the UAV is obtained based on the above-mentioned behavior consideration r rep The combined speed effect V generated by the speed of other drones within the range nv for:

[0161]

[0162] Among them, k nf k is the velocity alignment coefficient of the neighboring UAV, nv The speed influence coefficient of the UAV considering the speed of other friendly UAVs, V j is the speed of UAV j, r ij Represents drones i and r rep The distance between other drones j in the range, N d Indicates drone r rep the number of drones in range;

[0163] According to the repulsive speed effect of the friendly UAV based on the position threat and based on the above behavior considerations rep The comprehensive speed effect produced by the speed of other UAVs within the range is obtained, and the coordinated speed effect vector of the friendly UAV observation is obtained as follows:

[0164] V n =V np +V nv

[0165] Among them, V n Represents the coordinated velocity action vector of the friendly UAV observation.

[0166] In one embodiment, based on the vector sum of the group velocity action vector, the avoidance velocity action vector of the UAV avoiding the threat area, the tracking velocity action vector of the UAV tracking the mission target, and the observation and coordination velocity action vector of the neighboring UAV, as well as the UAV's motion constraints, the UAV controller is designed as follows:

[0167] V=V l ·R+(1-R)·(V e +V t +V d +V n )

[0168]

[0169] Among them, R is the inertial motion coefficient of the UAV, V max is the maximum speed of the drone under current conditions, V lThe speed of the drone at the last moment calculated by the controller.

[0170] For the specific definition of the distributed decision-making device for multi-objective drone swarms, please refer to the definition of the distributed decision-making method for multi-objective drone swarms above, which will not be repeated here. Each module in the above-mentioned distributed decision-making device for multi-objective drone swarms can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0171] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 13 As shown. The computer device includes a processor, memory, network interface, display screen and input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for controlling the motion of a distributed cluster of drones in a communication denial environment is implemented. The display screen of the computer device can be a liquid crystal display or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a key, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0172] Those skilled in the art will understand that Figure 13 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0173] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the method in the above embodiment when executing the computer program.

[0174] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method in the above embodiment are implemented.

[0175] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0176] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0177] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for controlling the motion of a distributed swarm of drones, characterized in that: The method comprises: Generate local environmental information about mission targets and threat areas in the current environment based on gene regulatory networks; Determine, based on the local environmental information concentration, a group speed action vector for the drones to maintain their group form, an avoidance speed action vector for the drones to avoid the threat area, a tracking speed action vector for the drones to track the mission target, and a friendly drone observation and coordination speed action vector; Design a controller for the drone based on the vector sum of the group velocity action vector, the avoidance velocity action vector for the drone to avoid the threat area, the tracking velocity action vector for the drone to track the mission target, and the observation and coordination velocity action vector of the neighboring drone, as well as the motion constraints of the drone; Use the designed controller to control the distributed swarm motion of UAVs; The controller of the UAV is designed based on the vector sum of the group velocity action vector, the UAV's avoidance velocity action vector for avoiding the threat area, the UAV's tracking velocity action vector for tracking the mission target, and the friendly UAV's observation and cooperation velocity action vector, as well as the UAV's motion constraints, including: According to the group velocity action vector , the evasion speed action vector of the UAV to avoid the threat area , the tracking speed action vector of the UAV tracking the mission target and the coordinated speed action vector of the friendly UAV observation The vector sum and the motion constraints of the UAV are used to design the controller of the UAV: ； ； in, is the UAV’s inertial motion coefficient, is the maximum flight speed of the drone under current conditions, The speed of the drone at the last moment calculated by the controller.

2. The method according to claim 1, characterized in that The steps to set the group velocity action vector for the drones to maintain the group form include: Set the group velocity action vector of the drones to maintain the group shape as: in, represents the group velocity action vector, represents an adjustable parameter, Indicates the concentration value of the drone position in the grid with the preset range centered on the drone’s location. Indicates the minimum concentration in the grid, The direction vector indicating the grid where the drone is located pointing to the grid with the minimum concentration.

3. The method according to claim 1, characterized in that The following are the action vectors for setting the evasion speed of the drone to avoid the threat area: The speed attenuation curve of the drone avoiding the threat area is obtained as follows: ； in, is the distance between the drone and the intended stopping point, is the acceleration parameter, is the linear gain; According to the speed attenuation curve, the evasion speed action vector of the UAV to avoid the threat area is obtained as follows: ； ； in, is the distance from the drone to its nearest obstacle, is the distance from the drone to its desired obstacle avoidance stopping point, and are the linear gain and acceleration parameters, represents the velocity vector of the virtual agent, Represents the components of the tracking velocity action vector.

4. The method according to claim 1, wherein The steps of setting the tracking speed action vector of the UAV to track the mission target include: Set the velocity action vector of the UAV tracking corresponding to the mission target to be: ； Represents the direction vector pointing from the UAV to the corresponding mission target; Set the velocity action vector of the drone to avoid collision with other mission targets: ； in, Represents the direction vector from the UAV to other UAVs in the grid; According to the speed action vector of the drone tracking the corresponding mission target and the speed action vector of the drone avoiding collision with other mission targets, the tracking speed action vector of the drone tracking the mission target is calculated as: 。 5. The method according to claim 1, wherein The steps for setting the coordinated velocity action vector for friendly drone observation include: According to the position of the friendly drone in the grid, set the repulsion speed effect of the friendly drone based on the position threat; According to the speed of the friendly drones in the grid, set the repulsive speed effect of the friendly drones based on speed threats; Set the speed alignment of neighboring drones based on the average speed vector of neighboring drones within the grid; According to the repulsive speed effect of friendly UAVs based on position threat, the repulsive speed effect of friendly UAVs based on speed threat and the speed alignment effect of friendly UAVs, the observation cooperative speed effect vector of friendly UAVs is obtained.

6. The method according to claim 5, characterized in that According to the position of the friendly drone within the grid, set the repulsion speed effect of the friendly drone based on the position threat, including: According to the position of the friendly UAV in the grid, the repulsion speed of the friendly UAV based on the position threat is set as follows: ； ； in, represents the repulsive velocity effect, Indicates drone and Other drones within range The distance between Consider the speed influence coefficient of other friendly UAVs for the UAV. Indicates drone Towards its neighbor drone The direction vector of .

7. The method according to claim 5, characterized in that According to the repulsive speed effect of the friendly UAV based on the position threat, the repulsive speed effect of the friendly UAV based on the speed threat, and the speed alignment effect of the friendly UAV, the friendly UAV observation cooperative speed effect vector is obtained, including: According to the speed-based repulsive speed effect of the friendly UAV and the speed alignment effect of the friendly UAV, the UAV is considered based on the behavior The combined speed effect of the speed of other drones within the range for: ； ； in, is the velocity alignment coefficient of the neighboring UAV, Consider the speed impact coefficient of other friendly UAVs for the UAV. For drones speed, Indicates drone and Other drones within range The distance between Indicates drone the number of drones in range; According to the repulsive speed effect of the friendly drone based on the position threat and based on the above behavior considerations The comprehensive speed effect generated by the speed of other UAVs within the range is obtained, and the coordinated speed effect vector of the friendly UAV observation is obtained as follows: ； in, Represents the coordinated velocity action vector of the friendly UAV observation.

Images