A method for swarm hunting control, an electronic device, and a storage medium

Abstract

The present application discloses a cluster hunting control method, an electronic device, and a storage medium, relating to the technical field of cluster control. The cluster hunting control method includes: obtaining the position information of a hunting target and an obstacle; then, generating a corresponding hunting mode through a gene regulatory network model according to the position information of the obstacle, where the hunting mode includes a squeezing type and a fusion type; and inputting the position information of the hunting target and the obstacle into the gene regulatory network model corresponding to the hunting mode to generate a comprehensive concentration field based on the position information of the hunting target and the obstacle. Then, determining a hunting form through the comprehensive concentration field; further, controlling the movement of an agent cluster through the gene regulatory network model according to the hunting form to achieve the hunting of the hunting target; thereby, obstacle avoidance conforming to the environmental characteristics can be realized, and further, the hunting efficiency is improved, and the success rate of the hunting task is further improved.

Description

Technical Field

[0001] This application relates to the technical field of cluster control, and particularly to a cluster hunting control method, an electronic device, and a storage medium. Background Art

[0002] Multi-robot systems (MRSs) are composed of a large number of small autonomous robots and are widely used in the execution of various tasks. The applications of such robot swarms include clustering, forming shapes, transporting large objects to targets, exploring unknown environments, and autonomously sequencing several specific tasks. The interactions between robots with each other and with the environment produce many desirable characteristics: scalability for different tasks, adaptability to harsh environments, and robustness to local damage. Generally speaking, swarm robots exhibit capabilities that are lacking in individual robots.

[0003] Using swarm robots to hunt targets is an emerging research hotspot. MRSs can be more suitable than humans for deployment in dangerous environments to achieve target hunting. However, when using swarm robots to hunt targets in related technologies, obstacle avoidance that conforms to the environmental characteristics cannot be achieved. Therefore, the hunting effect is affected (that is, it is difficult to achieve stable hunting performance), and even the hunting task may fail. Summary of the Invention

[0004] The main purpose of the embodiments of this application is to provide a cluster hunting control method, an electronic device, and a storage medium, aiming to improve the hunting efficiency.

[0005] To achieve the above objective, on the one hand, an embodiment of this application provides a cluster hunting control method, including: obtaining the position information of the hunting target and the obstacle; generating a corresponding hunting mode through a gene regulatory network model according to the position information of the obstacle, where the hunting mode includes a squeezing mode and a fusion mode; inputting the position information of the hunting target and the obstacle into the gene regulatory network model corresponding to the hunting mode to generate a comprehensive concentration field based on the position information of the hunting target and the obstacle, where the position information is determined by a local coordinate system; determining the hunting form through the comprehensive concentration field; controlling the movement of the intelligent agent cluster through the gene regulatory network model according to the hunting form to achieve the hunting of the hunting target; where the squeezing hunting mode is applied to the case where the positional relationship between obstacles is within a first set range, and the fusion hunting mode is applied to the case where the positional relationship between obstacles is within a second set range.

[0006] In some embodiments, the method for establishing the local coordinate system includes: determining a first agent, which is the agent closest to the encirclement target; setting the position where the first agent is located as the center of the local coordinate system; determining a second agent that is the closest to the first agent; determining the axis of the local coordinate system according to the positional relationship between the second agent and the first agent; determining a third agent, where the distances between the third agent, the first agent, and the second agent are less than a preset value; and determining the local coordinate system through the first agent, the second agent, and the third agent.

[0007] In some embodiments, generating a corresponding encirclement pattern through the gene regulatory network model according to the position information of the obstacle includes: determining the obstacle avoidance component velocity of the agent through the upper bound position information of the obstacle, the lower bound position information of the obstacle, the linear coefficient of the agent speed change, and the linear coefficient of the agent turning through the gene regulatory network model; determining the expected velocity of the agent by summing the obstacle avoidance component velocity of the agent and the encirclement task velocity in an obstacle-free environment; and determining the encirclement pattern through the gene regulatory network model based on the angle between the expected velocity and the encirclement task velocity in the obstacle-free environment.

[0008] In some embodiments, determining the obstacle avoidance component velocity of the agent through the upper bound position information of the obstacle, the lower bound position information of the obstacle, the linear coefficient of the agent speed change, and the linear coefficient of the agent turning through the gene regulatory network model includes: determining the obstacle avoidance component velocity of the agent through a first formula, and the first formula is expressed as:

[0009]

[0010] where, is the upper bound position information of the obstacle, is the lower bound position information of the obstacle, p f and p a respectively represent the linear coefficient of the agent speed change and the linear coefficient of the agent turning.

[0011] In some embodiments, generating a comprehensive concentration field based on the position information of the encirclement target and the obstacle includes: in the case where the encirclement pattern is the squeezing type, determining the comprehensive concentration field according to a second formula; the second formula is expressed as:

[0012]

[0013] Among them, p1 represents the position information of the surrounded target, and p2 represents the position information of the obstacle; g1 represents the concentration field formed according to the position information of the surrounded target and the obstacle; g2 represents the concentration field formed according to the position information of the obstacle; g3 represents the comprehensive concentration field that fuses g1 and g2, sig() represents the sigmod function, θ1, θ2, and θ3 are the thresholds of the s-shaped function, and k is the concentration field influence factor.

[0014] In some embodiments, generating the comprehensive concentration field based on the position information of the surrounded target and the obstacle further includes: in the case where the surrounding mode is the fusion type, determining the comprehensive concentration field according to the third formula; the third formula is expressed as:

[0015]

[0016] Among them, p1 represents the position information of the surrounded target, and p2 represents the position information of the obstacle; g1 represents the comprehensive concentration field formed according to g2 and the position information of the surrounded target and the obstacle; g2 represents the concentration field formed according to the position information of the obstacle; sig() represents the sigmod function, θ1, θ2, and θ3 are the thresholds of the s-shaped function, and k is the concentration field influence factor.

[0017] In some embodiments, determining the surrounding form through the comprehensive concentration field includes: projecting the comprehensive concentration field onto a two-dimensional plane to obtain a closed contour of concentration values; determining the surrounding form according to the closed contour.

[0018] In some embodiments, controlling the movement of the intelligent agent cluster through the gene regulatory network model according to the surrounding form to achieve the surrounding of the surrounded target includes: determining the movement speed of the leader in the cluster through the fourth formula, and the fourth formula is expressed as:

[0019] v il (t) = v is (t) + v it (t) + v io (t);

[0020] Among them, the leader is the intelligent agent that first senses the environmental anomaly information and adjusts its own movement mode; determining the movement speed of the follower in the cluster through the fifth formula, and the fifth formula is expressed as:

[0021] v if (t) = v is (t) + v it (t) + v io (t) + v iw (t);

[0022] Among them, v is (t) and v it (t) represent the amount of speed required to complete the basic encirclement task in a barrier-free environment, and v io (t) represents the obstacle avoidance component speed determined based on the encirclement pattern generated by the gene regulatory network model in an environment with obstacles, and v il (t) represents the movement speed of the leader in the cluster, and v if (t) represents the movement speed of the follower in the cluster, and v iw (t) represents the amount of speed used to accelerate the group's transformation of the encirclement form, and t represents time;

[0023]

[0024] Among them, v j (t) is the speed of neighbor agent j, and w j is the influence degree of neighbor agent j on agent i in accelerating the group's transformation of the encirclement form, and N is the number of neighbor agents within the detection range of agent i.

[0025] To achieve the above object, another aspect of the embodiments of the present application proposes an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above-mentioned cluster encirclement control method is implemented.

[0026] To achieve the above object, yet another aspect of the embodiments of the present application proposes a computer-readable storage medium, which stores a computer program. The computer program is characterized in that when the computer program is executed by a processor, the above-mentioned cluster encirclement control method is implemented.

[0027] The embodiments of the present application at least include the following beneficial effects:

[0028] The present application provides a cluster encirclement control method, an electronic device, and a storage medium. In the embodiments of the present application, first, the position information of the encirclement target and the obstacle is obtained; then, according to the position information of the obstacle, a corresponding encirclement pattern is generated through a gene regulatory network model, and the encirclement patterns include a squeezing type and a fusion type; and the position information of the encirclement target and the obstacle is input into the gene regulatory network model corresponding to the encirclement pattern to generate a comprehensive concentration field based on the position information of the encirclement target and the obstacle, wherein the position information is determined by a local coordinate system; then, the encirclement form is determined through the comprehensive concentration field; further, according to the encirclement form, the gene regulatory network model is used to control the movement of the agent cluster to achieve the encirclement of the encirclement target; wherein, the squeezing type encirclement pattern is applied to the case where the positional relationship between obstacles is within a first set range, and the fusion type encirclement pattern is applied to the case where the positional relationship between obstacles is within a second set range. In this way, obstacle avoidance conforming to the environmental characteristics can be achieved, thereby improving the encirclement efficiency and further increasing the success rate of the encirclement task.

[0029] Additional aspects and advantages of the present application will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0031] Figure 1 is a flowchart of the cluster encirclement control method provided by some embodiments of the present application;

[0032] Figure 2 is a schematic diagram of the local coordinate system provided by some embodiments of the present application;

[0033] Figure 3 is a schematic diagram of the squeezing type encirclement process in the cluster encirclement control method provided by some embodiments of the present application;

[0034] Figure 4 is a schematic diagram of the fusion type encirclement process in the cluster encirclement control method provided by some embodiments of the present application;

[0035] Figure 5 is a schematic diagram of the obstacle avoidance component velocity provided by some embodiments of the present application;

[0036] Figure 6 is a block diagram of the modules of the cluster encirclement control device provided by some embodiments of the present application;

[0037] Figure 7 is a schematic diagram of the hardware structure of the electronic device provided by some embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that when the term "embodiment" is mentioned herein, it means that the specific features, structures or characteristics described in connection with the embodiment may be included in at least one embodiment of the present application. The appearance of this phrase in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the embodiments of the present application. They are only examples of devices and methods that are consistent with some aspects of the embodiments of the present application as detailed in the appended claims.

[0039] It can be understood that the terms "first", "second", etc. used in the present application may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if", "when" as used herein may be interpreted as "when...", "when...", or "in response to determining".

[0040] The terms "at least one", "a plurality of", "each", "any one", etc. used in the present application, at least one includes one, two or more, a plurality includes two or more, each refers to each of the corresponding plurality, and any one refers to any one of the plurality.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0042] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0043] To make the inventive concept of this application easy to understand, before elaborating on the embodiments of this application in detail, the English abbreviations (terms) / related concepts involved in the embodiments of this application are first explained. The English abbreviations (terms) / related concepts involved in the embodiments of this application are applicable to the following explanations.

[0044] GRN: The full name of GRN is Gene Regulatory Network, which is interpreted in Chinese as gene regulatory network. A gene regulatory network refers to a network formed by the interaction relationships between genes within a cell or within a genome, specifically referring to gene regulation. This kind of network is a non-direct interaction relationship network established by regulating RNA or proteins. The research on gene regulatory networks mainly focuses on how to describe and analyze these complex networks to reveal the regulatory mechanisms of gene expression in organisms. In previous studies, there has already been the concept of using GRN and morphological gradients to achieve morphogenesis in swarm robot systems. The basic idea of applying the gene expression mechanism in biological morphogenesis to swarm robot control is to establish a metaphor between cells and robots. That is, each cell can be regarded as a robot. The genes in the cell generate proteins to construct a protein concentration field, which ultimately affects the movement of the cell in turn. Mapped to robots, a concentration field of target and obstacle information is established, and the concentration field generated by itself is used to regulate its own gene expression in turn, thereby regulating the life activities of the cell (i.e., the movement of the robot).

[0045] CH-GRN: CH-GRN is the Cooperative Hierarchical Gene Regulatory Network, which is a gene regulatory network applied in swarm robot systems, aiming to enhance the mutual cooperation between robots and utilize obstacles to achieve more effective and efficient target capture. The upper layer of this network design proposes a cooperation-based gene regulatory strategy to improve the overall performance of the system. CH-GRN realizes its functions through a multi-level gene regulatory network. In this network, the regulation of gene expression not only involves the regulation of a single gene, but also includes the interaction and synergistic effects between multiple genes.

[0046] Agent cluster: An agent cluster (Artificial Intelligence Entity Swarm) refers to a collection composed of multiple agents. These agents complete complex tasks and goals through collaborative work, mutual communication, and self-organization. The concept of agent clusters comes from swarm behaviors in nature, such as ant colonies, fish schools, and bird flocks, which achieve highly coordinated and efficient collective behaviors through simple individual behavior rules.

[0047] A swarm robotics system consists of a large number of small autonomous robots and is widely used for performing various tasks. Applications of such robot swarms include swarming, shape formation, transporting large objects to targets, exploring unknown environments, and autonomously sequencing several specific tasks. The interactions between robots with each other and with the environment give rise to many desirable characteristics: scalability for different tasks, adaptability to harsh environments, and robustness to local damage. Overall, swarm robots exhibit capabilities that are lacking in individual robots. In their applications, using swarm robots to surround a target is an emerging research hotspot. Swarm robotics systems can be more suitable than humans for deployment in dangerous environments to achieve target surround, especially in applications related to anti-terrorism, hazardous material marking in densely populated areas, and isolation of dangerous targets.

[0048] In related technologies, there is a method for forming a pattern of swarm agents based on an evolutionary hierarchical gene regulatory network model (EH-GRN), which improves the flexibility of pattern generation and the adaptability to various tasks. The upper layer of the gene regulatory network uses obstacle information and the position information of the surrounded target to generate environmental concentration information, and samples the concentration information to obtain a pattern (i.e., the surrounding shape) as the moving target of the agent. The lower layer of the gene regulatory network moves towards the target pattern while avoiding obstacles based on this environmental concentration information, and forms a surrounding circle near the surrounded target. The automatic generation of the upper layer structure of the gene regulatory network can be achieved by using gene programming methods. The gene network regulation model can be applicable to different application scenarios of swarm agents.

[0049] In related technologies, there is also a cooperative hierarchical gene regulatory network (CH-GRN), which aims to enhance the mutual cooperation between robot neighbors and use obstacles to achieve more effective and efficient capture. In the upper layer design of CH-GRN, a target-neighbor-obstacle (TNO) pattern generation method is proposed; it integrates the information of the target, neighbors, and obstacles to generate a more accurate pattern around the target. In the lower layer of CH-GRN, the concentration vector method is applied to enable the robot to quickly adapt to the pattern, thus completing the capture task.

[0050] However, in the related cluster surround technologies, when using obstacles for joint surround, the swarm robots cannot perform obstacle avoidance that conforms to the environmental characteristics, and a stable surround effect cannot be achieved, affecting the surround efficiency. In addition, CH-GRN only provides a cooperation framework, but the specific cooperation strategies need to be adjusted and optimized according to different tasks and environments.

[0051] In view of this, the present application proposes a cluster encirclement control method, an electronic device, and a storage medium. The solution first obtains the position information of the encirclement target and the obstacle. Then, according to the position information of the obstacle, a corresponding encirclement pattern is generated through a gene regulation network model. The encirclement patterns include a squeezing type and a fusion type. And the position information of the encirclement target and the obstacle is input into the gene regulation network model corresponding to the encirclement pattern to generate a comprehensive concentration field based on the position information of the encirclement target and the obstacle, where the position information is determined by a local coordinate system. Then, the encirclement form is determined through the comprehensive concentration field. Further, according to the encirclement form, the gene regulation network model is used to control the movement of the intelligent agent cluster to achieve the encirclement of the encirclement target. Among them, the squeezing type encirclement pattern is applied to the case where the positional relationship between obstacles is within a first set range, and the fusion type encirclement pattern is applied to the case where the positional relationship between obstacles is within a second set range. In this way, obstacle avoidance conforming to the environmental characteristics can be achieved, thereby improving the encirclement efficiency and further increasing the success rate of the encirclement task.

[0052] The method provided by the embodiments of the present application can be applied to the electronic device provided by the embodiments of the present application. Among them, the electronic device can be various types of intelligent agents or a shore-based device (a total control device).

[0053] The intelligent agent can be an intelligent robot, a drone, an unmanned ship, an unmanned vehicle, etc.

[0054] The shore-based device or the central control device can be a server, which can be an independent physical server, or a server cluster or a distributed system composed of multiple physical servers. It can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks, and big data and artificial intelligence platforms.

[0055] Next, the implementation steps of a cluster encirclement control method provided by the embodiments of the present application will be described in detail with reference to the drawings.

[0056] Please refer to Figure 1 , Figure 1 which is a flowchart of the cluster encirclement control method provided by some embodiments of the present application. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0057] The method of the embodiments of the present application includes the following steps:

[0058] Step 101: First, obtain the position information of the encirclement target and the obstacle;

[0059] Step 102: Then, based on the position information of the obstacles, a corresponding encirclement pattern is generated through the GRN (i.e., the gene regulatory network model). Among them, the generated encirclement patterns can include a squeezing encirclement pattern and a fusion encirclement pattern;

[0060] Step 103: Then, the position information of the encirclement target and the position information of the obstacles are input into the GRN corresponding to the encirclement pattern. Through the GRN, a comprehensive concentration field integrating the position information of the encirclement target and the obstacles can be generated. It should be understood that the position information is determined through a local coordinate system;

[0061] Step 104: Then, a suitable encirclement form can be determined according to the comprehensive concentration field;

[0062] Step 105: Finally, based on the determined encirclement form, the movement of the agent cluster can be controlled through the GRN to finally achieve the encirclement of the encirclement target;

[0063] The above squeezing encirclement pattern can be applied to the case where the positional relationship between the obstacles is within a first set range, and the above fusion encirclement pattern can be applied to the case where the positional relationship between the obstacles is within a second set range.

[0064] Steps 101 to 105 shown in the embodiments of the present application, by generating a corresponding encirclement pattern through the GRN with the position information of the obstacles, and inputting the position information of the encirclement target and the obstacles into the GRN corresponding to the encirclement pattern to generate a comprehensive concentration field based on the position information of the encirclement target and the obstacles, and determining the encirclement form through the comprehensive concentration field; further, controlling the movement of the agent cluster through the gene regulatory network model according to the encirclement form to achieve the encirclement of the encirclement target; thereby being able to achieve obstacle avoidance conforming to the environmental characteristics, and being able to introduce different types of GRNs to generate corresponding encirclement forms according to the characteristics of the environment. Under different environmental conditions, multiple agents can select the most suitable GRN model through an adaptive selection mechanism, so as to generate a specific encirclement pattern and achieve target encirclement.

[0065] The specific implementation methods of the above steps are introduced below.

[0066] In step 101, first, the position information of the encirclement target and the obstacles is obtained.

[0067] It should be understood that the position information of the obtained encirclement target and the position information of the obstacles are both obtained based on a local coordinate system, and the establishment of the local coordinate system will be introduced in the subsequent part.

[0068] The position information of the target to be surrounded can be the coordinate position of the target to be surrounded in the local coordinate system, and can also include the morphological information of the target to be surrounded, including the shape of the target to be surrounded, the size of the covered range, etc. This application does not limit this.

[0069] The position information of the obstacle can be the coordinate position of the obstacle in the local coordinate system, and can also include the shape information of the obstacle and the coverage range of the obstacle, or the shape formed by multiple obstacles and the specific range data, etc. Among them, the obstacle can include static obstacles or dynamic obstacles, actual obstacles and potential obstacles. This application does not limit this.

[0070] The acquisition of the position information can be through the sensors carried by the agent itself, such as lidar to obtain the corresponding position information. Among them, the position information of the obstacle and the target to be surrounded can be obtained by one or more agents. This application does not limit the method of obtaining the position information.

[0071] The embodiments provided in this application provide data support for generating the surrounding form through GRN subsequently by obtaining the position information of the target to be surrounded and the obstacle.

[0072] In some embodiments, the local coordinate system can be established by first determining the first agent, where the first agent can be one of the agents closest to the target to be surrounded; then, set the position where the first agent is located as the center of the local coordinate system (i.e., the origin, and the coordinate position can be set as (0,0)); then, set one of the agents closest to the first agent as the second agent; and set one of the axes of the local coordinate system according to the position relationship between the second agent and the first agent; then, the third agent can be determined. For the third agent, it is required that the distances between the third agent and the first agent and the second agent are all less than the preset value; finally, the local coordinate system (also called the local coordinate system) can be determined by the first agent, the second agent and the third agent.

[0073] It should be understood that setting the axis of the local coordinate system according to the position relationship between the second agent and the first agent can be setting the direction of the connection line between the second agent and the first agent as the horizontal axis of the local coordinate system, or setting the direction of the connection line between the two as the vertical axis of the local coordinate system. This application does not limit this.

[0074] The local coordinate system determined according to the above method can be a two-dimensional coordinate system or a three-dimensional coordinate system. This application does not limit this.

[0075] Among them, the agent can be an unmanned device or an intelligent robot. The unmanned device can be an unmanned vehicle, an unmanned aerial vehicle or an unmanned ship. This application does not limit this.

[0076] Exemplarily, referring to Figure 2 , Figure 2 which is a schematic diagram of a local coordinate system provided by some embodiments of the present application. Figure 2 A schematic of a two-dimensional plane local coordinate system is provided. The coordinate system includes an X-axis and a Y-axis. The origin of the coordinate system is not at the intersection of the X-axis and the Y-axis, but is located at the position of Center-bot. Herein, Center-bot represents the first intelligent agent, and the coordinate position of Center-bot in the local coordinate system is (0, 0). X-bot represents the second intelligent agent, and H-bot represents the third intelligent agent. In the figure, the red line segment represents the distance between the first intelligent agent and the second intelligent agent, the green line segment represents the distance between the second intelligent agent and the third intelligent agent, and the blue line segment represents the distance between the first intelligent agent and the third intelligent agent. Above, for the third intelligent agent, it is required that the distances between the third intelligent agent and the first and second intelligent agents pairwise are less than a preset value. That is to say, it is required that the length differences between the red line segment, the green line segment, and the blue line segment are less than the preset value. The preset value can be flexibly set according to the specific environment, and the present application does not limit this. Exemplarily, the length differences between the red line segment, the green line segment, and the blue line segment are 0, that is, the distances between the above intelligent agents are equal.

[0077] The horizontal axis of the local coordinate system is determined by the direction of the above red line segment, that is, the horizontal axis of the local coordinate system is parallel to the red line segment.

[0078] Exemplarily, the intelligent agent closest to the target can be determined as the center of the coordinate system, that is, the central robot. First, the central robot broadcasts the identification identifier (ID) of its nearest neighbor. The x-axis of the coordinate system is determined by the nearest neighbor, called X-bot. Once X-bot receives the message from the central robot and determines itself as X-bot, it sets its position to (d cx , 0), where d cx is the distance between X-bot and the central robot. Then, X-bot determines its position in the new coordinate system and broadcasts this position to other robots. Once the task is completed, the central robot can select a robot from those robots that can receive the information sent by X-bot as H-bot. As shown in Figure 2 . H-bot should meet the following condition, that is, H-bot, the central robot, and X-bot should maintain as equal distances as possible. Therefore, the relative position coordinates can be calculated as:

[0079]

[0080] where d hx is the distance between H-bot and X-bot, and d hcis the distance between the H-bot and the central robot, both of which are measured and calculated by the H-bot, d cx is the distance between the X-bot and the central robot. α is the angle between the H-bot and the x-axis. Other robots that can communicate with these three robots can calculate their relative positions after the three robots (central robot, X-bot, and H-bot) construct a local coordinate system.

[0081] In step 102, corresponding encirclement patterns can be generated based on the position information of the obstacles through the GRN (i.e., the gene regulatory network model). Among them, the generated encirclement patterns can include a squeezing encirclement pattern and a fusion encirclement pattern.

[0082] Among them, the position information of the obstacles can be detected by swarm robots. Exemplarily, each individual in the swarm robots can be equipped with a laser beam that can irradiate 360° counterclockwise from the current movement direction, and the distribution of obstacles in the environment can be detected by the situation where the laser beam is blocked.

[0083] In some embodiments, the position information of the obstacles can include the upper bound position information of the obstacles and the lower bound position information of the obstacles. Then, based on the determined position information above, the linear coefficient of the agent speed change, and the linear coefficient of the agent turning, the obstacle avoidance component speed of the agent is determined through the GRN.

[0084] Specifically, the obstacle avoidance component speed of the agent can be determined through the first formula, where the first formula can be expressed as:

[0085]

[0086] Among them, is the upper bound position information of the obstacle, is the lower bound position information of the obstacle, p f and p a are respectively used to represent the linear coefficient of the agent speed change and the linear coefficient of the agent turning.

[0087] Reference can be made to Figure 5 , Figure 5 is a schematic diagram of the obstacle avoidance component speed provided by some embodiments of the present application; Figure 5 In it, in the local coordinate system, the black dot represents one of the agents, and the filled gray box represents the detected obstacle. The upper bound position information of the obstacle can be detected by the agent and the lower bound position information Based on the detected upper bound position information and lower bound position information, combined with the linear coefficient of the agent speed change and the linear coefficient of the agent turning, the obstacle avoidance component speed v of the agent can be calculated i0It can be seen that the obstacle avoidance component velocity is in the direction opposite to that of the obstacle.

[0088] After that, the expected velocity of the robot (agent) is determined by the sum of the obstacle avoidance component velocity of the agent obtained through calculation and the pursuit task velocity in an obstacle-free environment.

[0089] Exemplarily, the expected velocity of the robot can be the final velocity of the agent at the next moment, expressed as:

[0090] v id (t) = v i (t) + v io (t);

[0091] where, v i (t) = v is (t) + v it (t), v is (t) and v it (t) are the velocity components that can ensure the completion of the basic pursuit task, i.e., the required velocity amount in an obstacle-free environment. t represents time, v i0 is the obstacle avoidance component velocity, and v id (t) is the final velocity of robot i at the next moment.

[0092] v it (t) = τ * (p i - p j );

[0093] p j is the coordinate (target coordinate) of robot j on the pursuit circle generated by the current GRN (i.e., the pursuit form or Pattern), and p i is the coordinate of the current robot i, and τ is the distance-dependent influence degree between robot i and the target.

[0094]

[0095] r esp is the obstacle avoidance safety distance, d ij is the distance between robot i and neighbor robot j, and a ij is the distance-dependent influence degree between robot i and neighbor robot j. N is the number of neighbor robots within the detection range of robot i, p i is the coordinate of robot i, and p j is the coordinate of robot j.

[0096] Then, through the expected velocity (the final velocity of the agent at the next moment) v id (t) and the pursuit task velocity v i(t) The included angle between determines the encirclement pattern through the upper network structure of the GRN.

[0097] Exemplarily, when the speed v id (t) and the encirclement task speed v in an obstacle-free environment i (t) The included angle between is greater than 90°, indicating that the current encirclement form conflicts with the obstacle avoidance requirements, and the encirclement form needs to be changed to adapt to the specific environment. That is, an encirclement pattern that can adapt to the characteristics of the environment around the target can be generated through the upper network structure of the GRN. Among them, the encirclement pattern can include a squeezing encirclement pattern and a fusion encirclement pattern. It should be understood that the realization of the final encirclement pattern is achieved by generating a comprehensive concentration field based on the position information of the encirclement target and the obstacle through the upper network structure of the GRN, and then determining the encirclement form.

[0098] In the embodiment provided by the present application, a corresponding encirclement pattern is generated through the upper network structure of the GRN according to the position information of the obstacle. Among them, the generated encirclement pattern can include a squeezing encirclement pattern and a fusion encirclement pattern. Since the morphological information of the obstacle is fully considered, the generated encirclement pattern can achieve the rapid completion of the encirclement task.

[0099] In step 103, the position information of the encirclement target and the position information of the obstacle are input into the GRN corresponding to the encirclement pattern. Through the GRN, a comprehensive concentration field integrating the position information of the encirclement target and the obstacle can be generated. It should be understood that the position information is determined through a local coordinate system.

[0100] The position information of the encirclement target and the position information of the obstacle are used together as the input of the gene regulatory network model to generate a comprehensive concentration field based on the position information of the encirclement target and the obstacle. It should be noted that in this concentration field, the concentration value at the position where the target is located is the highest, and the concentration values at other positions decrease as the spatial distance increases. When this comprehensive concentration field is mapped to a two-dimensional space, multiple closed isoclines will be formed around the target. According to the safe distances of the agents to the target and the obstacle, a suitable isocline is selected as the encirclement form.

[0101] The basic method for constructing the concentration field can be:

[0102] First, several targets and obstacles within the detection range of the agent are detected, and the calculation expression is as follows:

[0103]

[0104] Among them, p is the total matrix of the position information of all detected targets and obstacles; (T i,x , T i,y )(i = 1, 2,..., n) represents the position of the encirclement target, (O j,x , O j,y(j = 1, 2, …, m) represents the positions of the obstacles.

[0105] Secondly, all obstacle and target position information is regulated by the upper - layer network model of the GRN to form the final comprehensive concentration field. The calculation expression is as follows:

[0106]

[0107] g1 is responsible for regulating according to the position information to form the concentration field of the final required encirclement form; θ and k are adjustment parameters, which are the threshold of the sigmoid function and the concentration - field influence factor respectively, and p is the total matrix of all detected target and obstacle position information.

[0108] In some embodiments, a comprehensive concentration field based on the position information of the encirclement target and the obstacle is generated. Among them, in the case where the encirclement mode is the extrusion type, the corresponding comprehensive concentration field can be determined according to the second formula;

[0109] The second formula is expressed as:

[0110]

[0111]

[0112] Among them, p1 represents the position information of the encirclement target, p2 represents the position information of the obstacle; g1 represents the concentration field formed according to the position information of the encirclement target and the obstacle; g2 represents the concentration field formed according to the position information of the obstacle; g3 represents the comprehensive concentration field that fuses g1 and g2, sig() represents the sigmod function, θ1, θ2 and θ3 are the thresholds of the sigmoid function, and k is the concentration - field influence factor.

[0113] It should be understood that the upper - layer network structure of the GRN is responsible for generating an encirclement mode that can adapt to the characteristics of the target's surrounding environment. In this layer, two types of encirclement modes that can be adaptively adjusted are included, namely the extrusion - type encirclement mode and the fusion - type encirclement mode.

[0114] The extrusion type means that the GRN encirclement circle deforms under the influence of the obstacle, and this form is suitable for scenarios where the obstacle spacing is moderate. Exemplarily, please refer to Figure 3 , Figure 3 is a schematic diagram of the extrusion - type encirclement process in the cluster encirclement control method provided by some embodiments of the present application. Figure 3 In, the blue solid dots represent the agents Robot, the pentagram shape represents the encirclement target Target, the closed curve surrounded by the orange dotted line represents the encirclement form, i.e., Pattern, t represents time, and the gray rectangular shape represents the obstacle. In Figure 3In a relatively wide passageway enclosure scenario, it can be seen that the enclosure circle deforms at time t2, and at time t3, multiple agents can successfully complete the enclosure of the target in the wide passageway relying on this mode.

[0115] It should be understood that the above squeezing type enclosure mode can be applied to the case where the positional relationship between obstacles is within a first set range, and the above fusion type enclosure mode can be applied to the case where the positional relationship between obstacles is within a second set range. Among them, the first set range is usually larger than the second set range. Since the distance between obstacles satisfies the first set range, that is, the distance is still relatively large, therefore, the agent cluster can complete the pursuit of the enclosure target through squeezing deformation. And when the distance between obstacles satisfies the second set range, it means that the minimum distance between obstacles may be relatively small. At this time, the agent cluster cannot smoothly pass through the obstacles through squeezing deformation and at the same time achieve the pursuit of the enclosure target. Therefore, it is necessary to complete the pursuit of the obstacles through fusion deformation.

[0116] The fusion type can adaptively transform the enclosure form according to environmental conditions and use obstacles for auxiliary enclosure, and is suitable for scenarios where the obstacle intervals are relatively close or relatively special. Exemplarily, in Figure 4 as shown Figure 4 is a schematic diagram of the fusion type enclosure process in the cluster enclosure control method provided by some embodiments of the present application. Figure 4 Among them, the blue solid dots represent the agent Robot, the pentagram shape represents the enclosure target Target, the closed curve surrounded by the orange dotted line represents the enclosure form, that is, Pattern, t represents time, and the gray rectangular shape represents the obstacle. Figure 4 shows a narrow passageway enclosure scenario. If the agent cluster still adopts the squeezing type enclosure mode, it will face collision problems, so it will be transformed into the fusion type enclosure mode in the narrow passageway. As Figure 4 can be seen, the enclosure circle deforms at time t2, and at time t3, the multiple agent cluster can achieve the enclosure of the target with the help of obstacles.

[0117] In some embodiments, when the enclosure mode is the fusion type, the comprehensive concentration field can be determined according to the third formula;

[0118] The third formula is expressed as:

[0119]

[0120] Among them, p1 represents the position information of the target to be surrounded, and p2 represents the position information of the obstacle; g1 represents the comprehensive concentration field formed according to g2 and the position information of the target to be surrounded and the obstacle; g2 represents the concentration field formed according to the position information of the obstacle; sig() represents the sigmod function, θ1, θ2, and θ3 are the thresholds of the s-shaped function, and k is the concentration field influence factor.

[0121] In step 104, the applicable surrounding form can be determined according to the comprehensive concentration field.

[0122] In some embodiments, the comprehensive concentration field can be projected onto a two-dimensional plane, and then a closed contour line of the concentration value can be obtained; thereafter, the surrounding form can be determined according to the closed contour line.

[0123] When the comprehensive concentration field is mapped into a two-dimensional space, thus, multiple closed isoclines will be formed around the target to be surrounded. Thereafter, according to the safe distances of the agents to the target and the obstacle, a suitable isocline is selected as the surrounding form Pattern.

[0124] In step 105, based on the determined surrounding form, the agent swarm movement can be controlled by GRN to finally achieve the surrounding of the target to be surrounded.

[0125] It should be understood that the lower layer of GRN provides a complete group robot control model, which includes the basic conditions for driving the group robots to achieve surrounding and the self-adaptive mode conversion mechanism for assisting in achieving a better obstacle avoidance effect. In this layer, the group robots are also divided into emerging leaders and followers. The leader is the first object to sense the environmental anomaly, and the followers are the individuals in the group robots that sense the morphological change information of the leader. This method can achieve an accelerated change in the surrounding form.

[0126] The regulation mechanism of the lower layer can be divided into the regulation in a barrier-free environment and the regulation in an environment with obstacles. Among them, the regulation in a barrier-free environment has been introduced in the above-mentioned part of obtaining the obstacle avoidance component, and will not be elaborated here.

[0127] In an environment with obstacles, the motion model corresponding to the group robots can be determined by the fourth formula. Among them, the motion speed of the leader in the cluster (i.e., the individual in the cluster that first senses the environmental anomaly in the current environment and adjusts its own mode through the above-mentioned adaptive GRN mode conversion mechanism) can be expressed by the fourth formula as:

[0128] v il (t) = v is (t) + v it (t) + v io (t);

[0129] The motion speed of the followers in the cluster can be calculated by the fifth formula, and the fifth formula can be expressed as:

[0130] v if v(t) = v is (t) + v it (t) + v io (t) + v iw (t);

[0131] Wherein, v is (t) and v it (t) represent the speed required to complete the basic encirclement task in a barrier - free environment, v io (t) represents the obstacle - avoidance component speed determined based on the encirclement pattern generated by GRN in an environment with obstacles, v il (t) represents the movement speed of the leader in the swarm, v if (t) represents the movement speed of the follower in the swarm, v iw (t) represents the speed used to accelerate the swarm's transformation of the encirclement form, and t represents time;

[0132]

[0133] Wherein, v j (t) is the speed of neighbor agent j, w j is the influence degree of neighbor agent j on agent i in accelerating the swarm's transformation of the encirclement form, and N is the number of neighbor agents within the detection range of agent i.

[0134] By controlling the movement speed as described above, the swarm robots can move to the corresponding Pattern to achieve the encirclement of the encirclement target.

[0135] This application also provides an evaluation method for the encirclement performance, and the convergence error (C e ) can be used to measure the performance of the proposed method in the encirclement task: that is, the convergence error (C e ) represents the encirclement accuracy:

[0136]

[0137] Wherein, g = (g1, g2,..., g n ) T represents the swarm robots. f(g i ) is the implicit function for generating the encirclement pattern. For example, f(g i ): is a unit circle. d min (f(g i ), i, t) is the shortest distance between the i - th robot and the expected encirclement pattern at the t - th time step. C eZero indicates that all swarm robots are precisely distributed in the encirclement pattern. n represents the number of swarm robots. T represents the total number of time steps.

[0138] By setting a threshold for the convergence error, it can be determined whether the encirclement effect meets the expectations.

[0139] The above is an introduction to the swarm encirclement control method in the embodiments of this application.

[0140] Next, the implementation manners of the swarm encirclement control device provided in the embodiments of this application will be described in detail with reference to the accompanying drawings.

[0141] For the swarm encirclement control method provided in the above embodiments, the embodiments of this application also provide a swarm encirclement control device for implementing the above method, as Figure 6 shown Figure 6 is a schematic block diagram of the modules of the swarm encirclement control device in the embodiments of this application. The swarm encirclement control device includes:

[0142] An acquisition module, configured to first obtain the position information of the encirclement target and the obstacles;

[0143] A first generation module, configured to generate a corresponding encirclement pattern based on the position information of the obstacles through GRN (i.e., Gene Regulatory Network model), where the generated encirclement pattern may include a squeezing encirclement pattern and a fusion encirclement pattern;

[0144] A second generation module, configured to input the position information of the encirclement target and the position information of the obstacles into the GRN corresponding to the encirclement pattern, and through the GRN, a comprehensive concentration field integrating the position information of the encirclement target and the obstacles can be generated. It should be understood that the position information is determined through a local coordinate system;

[0145] A determination module, configured to determine an applicable encirclement form according to the comprehensive concentration field;

[0146] A control module, configured to control the movement of the intelligent agent swarm based on the determined encirclement form through GRN to finally achieve the encirclement of the encirclement target;

[0147] The above-mentioned squeezing encirclement pattern can be applied to the case where the positional relationship between the obstacles is within a first set range, and the above-mentioned fusion encirclement pattern can be applied to the case where the positional relationship between the obstacles is within a second set range.

[0148] It can be understood that the content in the above method embodiments is applicable to the device embodiments of this application. The functions specifically implemented by the device embodiments of this application are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those in the above method embodiments.

[0149] The embodiment of the present application further provides an electronic device, such as Figure 7 shown, the electronic device includes a memory, one or more processors ( Figure 7 only one is shown in the figure) and a computer program stored on the memory and executable on the processor. Among them: the memory is used to store software programs and units, and the processor executes various functional applications and data processing by running the software programs and units stored in the memory, so as to obtain the resources corresponding to the above preset events. Optionally, the processor realizes the above cluster hunting control method when running the above computer program stored in the memory.

[0150] As a non-transitory computer-readable medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely set relative to the processor, and these remote memories can be connected to the processor through a network.

[0151] It can be understood that the content in the above method embodiments is applicable to the embodiments of this electronic device. The functions specifically implemented by the embodiments of this electronic device are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0152] The embodiment of the present application further provides a computer program product. The above computer program product includes a computer program, and when the above computer program is executed by one or more processors, it can implement the steps of the above cluster hunting control method.

[0153] It can be understood that the content in the above method embodiments is applicable to the embodiments of this computer program product. The functions specifically implemented by the embodiments of this computer program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0154] For the cluster hunting control method, device, electronic device and computer program product provided by the embodiments of the present application, the intelligent agent can switch the adaptive hunting mode according to the characteristics of the environment. For example, when hunting a target in a channel, the squeezing or fusion hunting mode can be automatically switched according to the channel width. By introducing a variety of gene regulatory network models, a gene regulatory network model library is constructed. Under different environmental conditions, the most suitable model can be selected through an adaptive selection mechanism, so as to generate a corresponding hunting mode suitable for the environment.

[0155] In the embodiments of the present application, only the concentration information of obstacles within the detection range of the sensor is calculated, with a relatively small amount of calculation and low requirements for computing resources. In addition, multiple agents establish a local coordinate system through local communication without relying on global information.

[0156] The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art will know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0157] Although specific implementation schemes are described herein, those of ordinary skill in the art will recognize that many other modifications or alternative implementation schemes are also within the scope of the present disclosure. For example, any one of the functions and / or processing capabilities described in connection with a particular device or component can be performed by any other device or component. Additionally, although various illustrative specific implementations and architectures have been described in accordance with the embodiments of the present disclosure, those of ordinary skill in the art will recognize that many other modifications to the illustrative specific implementations and architectures herein are also within the scope of the present disclosure.

[0158] Certain aspects of the present disclosure have been described above with reference to block diagrams and flowcharts of systems, methods, systems, and / or computer program products according to exemplary embodiments. It should be understood that one or more blocks in the block diagrams and flowcharts, and combinations of blocks in the block diagrams and flowcharts, can be implemented respectively by executing computer-executable program instructions. Similarly, according to some embodiments, some blocks in the block diagrams and flowcharts may not need to be executed in the order shown, or may not need to be executed at all. Additionally, additional components and / or operations beyond those shown in the blocks of the block diagrams and flowcharts may exist in certain embodiments.

[0159] Therefore, the blocks in the block diagrams and flowcharts support combinations of devices for performing the specified functions, combinations of elements or steps for performing the specified functions, and program instruction devices for performing the specified functions. It should also be understood that each block in the block diagrams and flowcharts, and combinations of blocks in the block diagrams and flowcharts, can be implemented by a dedicated hardware computer system that performs a specific function, element, or step, or a combination of dedicated hardware and computer instructions.

[0160] The program modules, applications, etc. described herein may include one or more software components, including, for example, software objects, methods, data structures, etc. Each such software component may include computer-executable instructions that, upon execution, cause at least a portion of the functions described herein (e.g., one or more operations of the illustrative methods described herein) to be performed.

[0161] Software components can be coded in any of a variety of programming languages. An exemplary programming language can be a low-level programming language, such as an assembly language associated with a particular hardware architecture and / or operating system platform. Software components including assembly language instructions may need to be converted by an assembler into executable machine code before being executed by the hardware architecture and / or platform. Another exemplary programming language can be a higher-level programming language, which can be portable across multiple architectures. Software components including a higher-level programming language may need to be converted by an interpreter or compiler into an intermediate representation before execution. Other examples of programming languages include, but are not limited to, macro languages, shell or command languages, job control languages, scripting languages, database query or search languages, or report writing languages. In one or more exemplary embodiments, software components containing instructions in one of the above examples of programming languages can be directly executed by an operating system or other software components without first being converted into another form.

[0162] Software components can be stored as files or other data storage constructs. Software components having similar types or related functions can be stored together in, for example, a particular directory, folder, or library. Software components can be static (e.g., pre-set or fixed) or dynamic (e.g., created or modified at execution time).

[0163] The embodiments of the present application have been described in detail above with reference to the accompanying drawings. However, the present application is not limited to the above embodiments, and various changes can be made without departing from the spirit of the present application within the knowledge scope of those of ordinary skill in the art.

Claims

1. A method for controlling collective hunting, characterized in that Including the following steps: Obtain the position information of the target to be surrounded and the obstacles; Generate corresponding surrounding patterns through a gene regulatory network model according to the position information of the obstacles, where the surrounding patterns include a squeezing type and a fusion type; Input the position information of the target to be surrounded and the obstacles into the gene regulatory network model corresponding to the surrounding pattern to generate a comprehensive concentration field based on the position information of the target to be surrounded and the obstacles, where the position information is determined by a local coordinate system; Determine the surrounding form through the comprehensive concentration field; Control the movement of the agent cluster through the gene regulatory network model according to the surrounding form to achieve the surrounding of the target to be surrounded; Among them, the squeezing surrounding pattern is applied to the case where the positional relationship between the obstacles is within a first set range, and the fusion surrounding pattern is applied to the case where the positional relationship between the obstacles is within a second set range; Among them, the generating of the corresponding surrounding pattern through the gene regulatory network model according to the position information of the obstacles includes: Determine the obstacle avoidance component speed of the agent through the upper bound position information of the obstacle, the lower bound position information of the obstacle, the linear coefficient of the agent speed change, and the linear coefficient of the agent turning through the gene regulatory network model; Determine the expected speed of the agent by the sum of the obstacle avoidance component speed of the agent and the surrounding task speed in an obstacle-free environment; Determine the surrounding pattern through the gene regulatory network model by the angle between the expected speed and the surrounding task speed in an obstacle-free environment; The determining of the surrounding form through the comprehensive concentration field includes: Project the comprehensive concentration field onto a two-dimensional plane to obtain a closed contour line of the concentration value; Determine the surrounding form according to the closed contour line; 2. The cluster round-up control method according to claim 1, characterized in that, The method for establishing the local coordinate system includes: Determine a first agent, where the first agent is the agent closest to the target to be surrounded; Set the position where the first agent is located as the center of the local coordinate system; Determine the agent closest to the first agent as the second agent; Determine the axis of the local coordinate system according to the positional relationship between the second agent and the first agent; Determine a third agent, where the distance between the third agent, the first agent, and the second agent is less than a preset value; Determine the local coordinate system through the first agent, the second agent, and the third agent; 3. The cluster round-up control method according to claim 1, characterized in that, The determining of the obstacle avoidance component speed of the agent through the upper bound position information of the obstacle, the lower bound position information of the obstacle, the linear coefficient of the agent speed change, and the linear coefficient of the agent turning through the gene regulatory network model includes: Determine the obstacle avoidance component speed of the agent through a first formula, and the first formula is expressed as: Among them, is the upper bound position information of the obstacle, is the lower bound position information of the obstacle, p f and p a respectively represent the linear coefficient of the speed change of the agent and the linear coefficient of the turning of the agent.

4. The cluster round-up control method according to claim 1, wherein The generating of the comprehensive concentration field based on the position information of the target to be surrounded and the obstacles includes: In the case where the surrounding pattern is the squeezing type, determine the comprehensive concentration field according to a second formula; The second formula is expressed as: Among them, p1 represents the position information of the surrounded target, and p2 represents the position information of the obstacle; g1 represents the concentration field formed according to the position information of the surrounded target and the obstacle; g2 represents the concentration field formed according to the position information of the obstacle; g3 represents the comprehensive concentration field that fuses g1 and g2, sig() represents the sigmod function, θ1, θ2, and θ3 are the thresholds of the s-shaped function, and k is the concentration field influence factor.

5. The cluster roundup control method according to claim 1, wherein The generation of the comprehensive concentration field based on the position information of the surrounded target and the obstacle further includes: When the surrounded mode is the fusion type, determining the comprehensive concentration field according to the third formula; The third formula is expressed as: Among them, p1 represents the position information of the surrounded target, and p2 represents the position information of the obstacle; g1 represents the comprehensive concentration field formed according to g2 and the position information of the surrounded target and the obstacle; g2 represents the concentration field formed according to the position information of the obstacle; sig() represents the sigmod function, θ1, θ2, and θ3 are the thresholds of the s-shaped function, and k is the concentration field influence factor.

6. The cluster rounding-up control method according to claim 1, characterized in that, The control of the movement of the intelligent agent cluster through the gene regulatory network model according to the surrounded form to achieve the surrounding of the surrounded target includes: Determining the movement speed of the leader in the cluster through the fourth formula, and the fourth formula is expressed as: v il v(t) = v is v(t) + v it v(t) + v io v(t); Among them, the leader is the intelligent agent that first perceives the environmental anomaly information and adjusts its own movement mode; Determining the movement speed of the follower in the cluster through the fifth formula, and the fifth formula is expressed as: v if v(t) = v is v(t) + v it v(t) + v io v(t) + v iw v(t); Among them, v is (t) and v it (t) represent the amount of speed required to complete the basic encirclement task in a barrier-free environment, v io (t) represents the obstacle avoidance component speed determined based on the encirclement pattern generated by the gene regulatory network model in an environment with obstacles, v il (t) represents the movement speed of the leader in the cluster, v if (t) represents the movement speed of the follower in the cluster, v iw (t) represents the amount of speed used to accelerate the group's transformation of the encirclement form, and t represents time; Among them, v j (t) is the velocity of neighbor agent j, w j is the influence degree of neighbor agent j on agent i in accelerating the group transformation to the encirclement and pursuit form, and N is the number of neighbor agents within the detection range of agent i.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the cluster surrounding control method according to any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the cluster surrounding control method according to any one of claims 1 to 6.

Images